Datasets:

---

ubowang

/

cite-llm-multi-cite-eval

like 0

[ "<|reference_start|> DATFNets-dynamic adaptive assigned transformer network for fire detection: <|reference_end|>", "<|reference_start|> Early Stage Fire Detection System Based on Shallow Guide Deep Network: <|reference_end|>", "<|reference_start|> An Improved YOLOv5s Fire Detection Model: <|reference_end|>", "<|reference_start|> FS-YOLO: a multi-scale SAR ship detection network in complex scenes: <|reference_end|>" ]

[ 17, 21, 36, 37 ]

[ "<|reference_start|> A QVGA 143\ndB Dynamic Range Frame-Free PWM Image Sensor With\nLossless Pixel-Level Video Compression and Time-Domain\nCDS: The biomimetic CMOS dynamic vision and image sensor described in this paper is based on a QVGA (304×240) array of fully autonomous pixels containing event-based change detection and pulse-width-modulation (PWM) imaging circuitry. Exposure measurements are initiated and carried out locally by the individual pixel that has detected a change of brightness in its field-of-view. Pixels do not rely on external timing signals and independently and asynchronously request access to an (asynchronous arbitrated) output channel when they have new grayscale values to communicate. Pixels that are not stimulated visually do not produce output. The visual information acquired from the scene, temporal contrast and grayscale data, are communicated in the form of asynchronous address-events (AER), with the grayscale values being encoded in inter-event intervals. The pixel-autonomous and massively parallel operation ideally results in lossless video compression through complete temporal redundancy suppression at the pixel level. Compression factors depend on scene activity and peak at ~1000 for static scenes. Due to the time-based encoding of the illumination information, very high dynamic range - intra-scene DR of 143 dB static and 125 dB at 30 fps equivalent temporal resolution - is achieved. A novel time-domain correlated double sampling (TCDS) method yields array FPN of <;0.25% rms. SNR is >56 dB (9.3 bit) for >10 Lx illuminance. <|reference_end|>", "<|reference_start|> EVO: A geometric approach to event-based 6-dof parallel tracking and mapping in real time: We present EVO, an event-based visual odometry algorithm. Our algorithm successfully leverages the outstanding properties of event cameras to track fast camera motions while recovering a semidense three-dimensional (3-D) map of the environment. The implementation runs in real time on a standard CPU and outputs up to several hundred pose estimates per second. Due to the nature of event cameras, our algorithm is unaffected by motion blur and operates very well in challenging, high dynamic range conditions with strong illumination changes. To achieve this, we combine a novel, event-based tracking approach based on image-to-model alignment with a recent event-based 3-D reconstruction algorithm in a parallel fashion. Additionally, we show that the output of our pipeline can be used to reconstruct intensity images from the binary event stream, though our algorithm does not require such intensity information. We believe that this work makes significant progress in simultaneous localization and mapping by unlocking the potential of event cameras. This allows us to tackle challenging scenarios that are currently inaccessible to standard cameras. <|reference_end|>", "<|reference_start|> A Low Power, High Throughput, Fully Event-Based Stereo\nSystem: We introduce a stereo correspondence system implemented fully on event-based digital hardware, using a fully graph-based non von-Neumann computation model, where no frames, arrays, or any other such data-structures are used. This is the first time that an end-to-end stereo pipeline from image acquisition and rectification, multi-scale spatiotemporal stereo correspondence, winner-take-all, to disparity regularization is implemented fully on event-based hardware. Using a cluster of TrueNorth neurosynaptic processors, we demonstrate their ability to process bilateral event-based inputs streamed live by Dynamic Vision Sensors (DVS), at up to 2,000 disparity maps per second, producing high fidelity disparities which are in turn used to reconstruct, at low power, the depth of events produced from rapidly changing scenes. Experiments on real-world sequences demonstrate the ability of the system to take full advantage of the asynchronous and sparse nature of DVS sensors for low power depth reconstruction, in environments where conventional frame-based cameras connected to synchronous processors would be inefficient for rapidly moving objects. System evaluation on event-based sequences demonstrates a ~ 200 × improvement in terms of power per pixel per disparity map compared to the closest state-of-the-art, and maximum latencies of up to 11ms from spike injection to disparity map ejection. <|reference_end|>", "<|reference_start|> A combined corner and edge detector: The problem we are addressing in Alvey Project MMI149 is that of using computer vision to understand the unconstrained 3D world, in which the viewed scenes will in general contain too wide a diversity of objects for topdown recognition techniques to work. For example, we desire to obtain an understanding of natural scenes, containing roads, buildings, trees, bushes, etc., as typified by the two frames from a sequence illustrated in Figure 1. The solution to this problem that we are pursuing is to use a computer vision system based upon motion analysis of a monocular image sequence from a mobile camera. By extraction and tracking of image features, representations of the 3D analogues of these features can be constructed. <|reference_end|>" ]

[ 4, 8, 9, 12 ]

[ "<|reference_start|> Coordinating Multiple Antenna Cellular Networks to Achieve Enormous Spectral Efficiency: Intercell interference limits the capacity of wireless networks. To mitigate this interference we explore coherently coordinated transmission (CCT) from multiple base stations to each user. To treat users fairly, we explore equal rate (ER) networks. We evaluate the downlink network efficiency of CCT as compared to serving each user with single base transmission (SBT) with a separate base uniquely assigned to each user. Efficiency of ER networks is measured as total network throughput relative to the number of network antennas at 10% user outage. Efficiency is compared relative to the baseline of single base transmission with power control, (ER-SBT), where base antenna transmissions are not coordinated and apart from power control and the assignment of 10% of the users to outage, nothing is done to mitigate interference. We control the transmit power of ER systems to maximise the common rate for ER-SBT, ER-CCT based on zero forcing, and ER-CCT employing dirty paper coding. We do so for (no. of transmit antennas per base, no. of receive antennas per user) equal to (1,1), (2,2) and (4,4). We observe that CCT mutes intercell interference enough, so that enormous spectral efficiency improvement associated with using multiple antennas in isolated communication links occurs as well for the base-to-user links in a cellular network. <|reference_end|>", "<|reference_start|> {An iteratively weighted MMSE approach to distributed sum-utility maximization for a MIMO interfering broadcast channel: Consider the MIMO interfering broadcast channel whereby multiple base stations in a cellular network simultaneously transmit signals to a group of users in their own cells while causing interference to the users in other cells. The basic problem is to design linear beamformers that can maximize the system throughput. In this paper we propose a linear transceiver design algorithm for weighted sum-rate maximization that is based on iterative minimization of weighted mean squared error (MSE). The proposed algorithm only needs local channel knowledge and converges to a stationary point of the weighted sum-rate maximization problem. Furthermore, we extend the algorithm to a general class of utility functions and establish its convergence. The resulting algorithm can be implemented in a distributed asynchronous manner. The effectiveness of the proposed algorithm is validated by numerical experiments. <|reference_end|>", "<|reference_start|> Joint Beamforming and Power Control in Coordinated Multicell: Max-Min Duality, Effective Network and Large System Transition: This paper studies joint beamforming and power control in a coordinated multicell downlink system that serves multiple users per cell to maximize the minimum weighted signal-to-interference-plus-noise ratio. The optimal solution and distributed algorithm with geometrically fast convergence rate are derived by employing the nonlinear Perron-Frobenius theory and the multicell network duality. The iterative algorithm, though operating in a distributed manner, still requires instantaneous power update within the coordinated cluster through the backhaul. The backhaul information exchange and message passing may become prohibitive with increasing number of transmit antennas and increasing number of users. In order to derive asymptotically optimal solution, random matrix theory is leveraged to design a distributed algorithm that only requires statistical information. The advantage of our approach is that there is no instantaneous power update through backhaul. Moreover, by using nonlinear Perron-Frobenius theory and random matrix theory, an effective primal network and an effective dual network are proposed to characterize and interpret the asymptotic solution. <|reference_end|>", "<|reference_start|> FemtoCaching: Wireless Video Content Delivery through Distributed Caching Helpers: Video on-demand streaming from Internet-based servers is becoming one of the most important services offered by wireless networks today. In order to improve the area spectral efficiency of video transmission in cellular systems, small cells heterogeneous architectures (e.g., femtocells, WiFi off-loading) are being proposed, such that video traffic to nomadic users can be handled by short-range links to the nearest small cell access points (referred to as \"helpers\"). As the helper deployment density increases, the backhaul capacity becomes the system bottleneck. In order to alleviate such bottleneck we propose a system where helpers with low-rate backhaul but high storage capacity cache popular video files. Files not available from helpers are transmitted by the cellular base station. We analyze the optimum way of assigning files to the helpers, in order to minimize the expected downloading time for files. We distinguish between the uncoded case (where only complete files are stored) and the coded case, where segments of Fountain-encoded versions of the video files are stored at helpers. We show that the uncoded optimum file assignment is NP-hard, and develop a greedy strategy that is provably within a factor 2 of the optimum. Further, for a special case we provide an efficient algorithm achieving a provably better approximation ratio of $1-(1-1/d)^d$, where $d$ is the maximum number of helpers a user can be connected to. We also show that the coded optimum cache assignment problem is convex that can be further reduced to a linear program. We present numerical results comparing the proposed schemes. <|reference_end|>" ]

[ 2, 3, 8, 13 ]

[ "<|reference_start|> A Syntactic Neural Model for General-Purpose Code Generation: We consider the problem of parsing natural language descriptions into source code written in a general-purpose programming language like Python. Existing data-driven methods treat this problem as a language generation task without considering the underlying syntax of the target programming language. Informed by previous work in semantic parsing, in this paper we propose a novel neural architecture powered by a grammar model to explicitly capture the target syntax as prior knowledge. Experiments find this an effective way to scale up to generation of complex programs from natural language descriptions, achieving state-of-the-art results that well outperform previous code generation and semantic parsing approaches. <|reference_end|>", "<|reference_start|> Abstract Syntax Networks for Code Generation and Semantic Parsing: Tasks like code generation and semantic parsing require mapping unstructured (or partially structured) inputs to well-formed, executable outputs. We introduce abstract syntax networks, a modeling framework for these problems. The outputs are represented as abstract syntax trees (ASTs) and constructed by a decoder with a dynamically-determined modular structure paralleling the structure of the output tree. On the benchmark Hearthstone dataset for code generation, our model obtains 79.2 BLEU and 22.7% exact match accuracy, compared to previous state-of-the-art values of 67.1 and 6.1%. Furthermore, we perform competitively on the Atis, Jobs, and Geo semantic parsing datasets with no task-specific engineering. <|reference_end|>", "<|reference_start|> CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation: Pre-trained models for Natural Languages (NL) like BERT and GPT have been recently shown to transfer well to Programming Languages (PL) and largely benefit a broad set of code-related tasks. Despite their success, most current methods either rely on an encoder-only (or decoder-only) pre-training that is suboptimal for generation (resp. understanding) tasks or process the code snippet in the same way as NL, neglecting the special characteristics of PL such as token types. We present CodeT5, a unified pre-trained encoder-decoder Transformer model that better leverages the code semantics conveyed from the developer-assigned identifiers. Our model employs a unified framework to seamlessly support both code understanding and generation tasks and allows for multi-task learning. Besides, we propose a novel identifier-aware pre-training task that enables the model to distinguish which code tokens are identifiers and to recover them when they are masked. Furthermore, we propose to exploit the user-written code comments with a bimodal dual generation task for better NL-PL alignment. Comprehensive experiments show that CodeT5 significantly outperforms prior methods on understanding tasks such as code defect detection and clone detection, and generation tasks across various directions including PL-NL, NL-PL, and PL-PL. Further analysis reveals that our model can better capture semantic information from code. Our code and pre-trained models are released at https: //github.com/salesforce/CodeT5 . <|reference_end|>", "<|reference_start|> Evaluating Large Language Models Trained on Code: We introduce Codex, a GPT language model fine-tuned on publicly available code from GitHub, and study its Python code-writing capabilities. A distinct production version of Codex powers GitHub Copilot. On HumanEval, a new evaluation set we release to measure functional correctness for synthesizing programs from docstrings, our model solves 28.8% of the problems, while GPT-3 solves 0% and GPT-J solves 11.4%. Furthermore, we find that repeated sampling from the model is a surprisingly effective strategy for producing working solutions to difficult prompts. Using this method, we solve 70.2% of our problems with 100 samples per problem. Careful investigation of our model reveals its limitations, including difficulty with docstrings describing long chains of operations and with binding operations to variables. Finally, we discuss the potential broader impacts of deploying powerful code generation technologies, covering safety, security, and economics. <|reference_end|>" ]

[ 0, 1, 7, 10 ]

[ "<|reference_start|> Scalability! But at what {COST}?: We offer a new metric for big data platforms, COST, or the Configuration that Outperforms a Single Thread. The COST of a given platform for a given problem is the hardware configuration required before the platform outperforms a competent single-threaded implementation. COST weighs a system's scalability against the overheads introduced by the system, and indicates the actual performance gains of the system, without rewarding systems that bring substantial but parallelizable overheads. \n \nWe survey measurements of data-parallel systems recently reported in SOSP and OSDI, and find that many systems have either a surprisingly large COST, often hundreds of cores, or simply underperform one thread for all of their reported configurations. <|reference_end|>", "<|reference_start|> mmap: Fast billion-scale graph computation on a pc via memory mapping: Graph computation approaches such as GraphChi and TurboGraph recently demonstrated that a single PC can perform efficient computation on billion-node graphs. To achieve high speed and scalability, they often need sophisticated data structures and memory management strategies. We propose a minimalist approach that forgoes such requirements, by leveraging the fundamental memory mapping (MMap) capability found on operating systems. We contribute: (1) a new insight that MMap is a viable technique for creating fast and scalable graph algorithms that surpasses some of the best techniques; (2) the design and implementation of popular graph algorithms for billion-scale graphs with little code, thanks to memory mapping; (3) extensive experiments on real graphs, including the 6.6 billion edge Yahoo Web graph, and show that this new approach is significantly faster or comparable to the highly-optimized methods (e.g., 9.5X faster than GraphChi for computing PageRank on 1.47B edge Twitter graph). We believe our work provides a new direction in the design and development of scalable algorithms. Our packaged code is available at http://poloclub.gatech.edu/mmap/. <|reference_end|>", "<|reference_start|> mmap: Fast billion-scale graph computation on a pc via memory mapping: Graph computation approaches such as GraphChi and TurboGraph recently demonstrated that a single PC can perform efficient computation on billion-node graphs. To achieve high speed and scalability, they often need sophisticated data structures and memory management strategies. We propose a minimalist approach that forgoes such requirements, by leveraging the fundamental memory mapping (MMap) capability found on operating systems. We contribute: (1) a new insight that MMap is a viable technique for creating fast and scalable graph algorithms that surpasses some of the best techniques; (2) the design and implementation of popular graph algorithms for billion-scale graphs with little code, thanks to memory mapping; (3) extensive experiments on real graphs, including the 6.6 billion edge Yahoo Web graph, and show that this new approach is significantly faster or comparable to the highly-optimized methods (e.g., 9.5X faster than GraphChi for computing PageRank on 1.47B edge Twitter graph). We believe our work provides a new direction in the design and development of scalable algorithms. Our packaged code is available at http://poloclub.gatech.edu/mmap/. <|reference_end|>", "<|reference_start|> Scalability! But at what {COST}?: We offer a new metric for big data platforms, COST, or the Configuration that Outperforms a Single Thread. The COST of a given platform for a given problem is the hardware configuration required before the platform outperforms a competent single-threaded implementation. COST weighs a system's scalability against the overheads introduced by the system, and indicates the actual performance gains of the system, without rewarding systems that bring substantial but parallelizable overheads. \n \nWe survey measurements of data-parallel systems recently reported in SOSP and OSDI, and find that many systems have either a surprisingly large COST, often hundreds of cores, or simply underperform one thread for all of their reported configurations. <|reference_end|>" ]

[ 0, 1, 2, 3 ]

[ "<|reference_start|> Determining the freshness of fruits in the food industry by image classification using transfer learning: <|reference_end|>", "<|reference_start|> Automated apple defect detection using state-of-the-art object detection techniques: <|reference_end|>", "<|reference_start|> Automated apple defect detection using state-of-the-art object detection techniques: <|reference_end|>", "<|reference_start|> A Simple and Efficient Deep Learning-Based Framework for Automatic Fruit\nRecognition: Accurate detection and recognition of various kinds of fruits and vegetables by using the artificial intelligence (AI) approach always remain a challenging task due to similarity between various types of fruits and challenging environments such as lighting and background variations. Therefore, developing and exploring an expert system for automatic fruits' recognition is getting more and more important after many successful approaches; however, this technology is still far from being mature. The deep learning-based models have emerged as state-of-the-art techniques for image segmentation and classification and have a lot of promise in challenging domains such as agriculture, where they can deal with the large variability in data better than classical computer vision methods. In this study, we proposed a deep learning-based framework to detect and recognize fruits and vegetables automatically with difficult real-world scenarios. The proposed method might be helpful for the fruit sellers to identify and differentiate various kinds of fruits and vegetables that have similarities. The proposed method has applied deep convolutional neural network (DCNN) to the undertakings of distinguishing natural fruit images of the Gilgit-Baltistan (GB) region as this area is famous for fruits' production in Pakistan as well as in the world. The experimental outcomes demonstrate that the suggested deep learning algorithm has the effective capability of automatically recognizing the fruit with high accuracy of 96%. This high accuracy exhibits that the proposed approach can meet world application requirements. <|reference_end|>" ]

[ 7, 8, 12, 13 ]

[ "<|reference_start|> Handbook of Weighted Automata: <|reference_end|>", "<|reference_start|> Weighted Finite-State Transducer Algorithms. An Overview: <|reference_end|>", "<|reference_start|> Active learning of nondeterministic finite state machines: We consider the problem of learning nondeterministic finite state machines (NFSMs) from systems where their internal structures are implicit and nondeterministic. Recently, an algorithm for inferring observable NFSMs (ONFSMs), which are the potentially learnable subclass of NFSMs, has been proposed based on the hypothesis that the complete testing assumption is satisfied. According to this assumption, with an input sequence (query), the complete set of all possible output sequences is given by the so-called Teacher, so the number of times for asking the same query is not taken into account in the algorithm. In this paper, we propose , a refined ONFSM learning algorithm that considers the amount for repeating the same query as one parameter. Unlike the previous work, our approach does not require all possible output sequences in one answer. Instead, it tries to observe the possible output sequences by asking the same query many times to the Teacher. We have proved that can infer the corresponding ONFSMs of the unknown systems when the number of tries for the same query is adequate to guarantee the complete testing assumption. Moreover, the proof shows that our algorithm will eventually terminate no matter whether the assumption is fulfilled or not. We also present the theoretical time complexity analysis of . In addition, experimental results demonstrate the practical efficiency of our approach. <|reference_end|>", "<|reference_start|> INFERRING REGULAR LANGUAGES IN POLYNOMIAL UPDATED TIME: <|reference_end|>" ]

[ 0, 2, 5, 7 ]

[ "<|reference_start|> The loss surface of deep and wide neural networks: While the optimization problem behind deep neural networks is highly non-convex, it is frequently observed in practice that training deep networks seems possible without getting stuck in suboptimal points. It has been argued that this is the case as all local minima are close to being globally optimal. We show that this is (almost) true, in fact almost all local minima are globally optimal, for a fully connected network with squared loss and analytic activation function given that the number of hidden units of one layer of the network is larger than the number of training points and the network structure from this layer on is pyramidal. <|reference_end|>", "<|reference_start|> On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima: The stochastic gradient descent (SGD) method and its variants are algorithms of choice for many Deep Learning tasks. These methods operate in a small-batch regime wherein a fraction of the training data, say $32$-$512$ data points, is sampled to compute an approximation to the gradient. It has been observed in practice that when using a larger batch there is a degradation in the quality of the model, as measured by its ability to generalize. We investigate the cause for this generalization drop in the large-batch regime and present numerical evidence that supports the view that large-batch methods tend to converge to sharp minimizers of the training and testing functions - and as is well known, sharp minima lead to poorer generalization. In contrast, small-batch methods consistently converge to flat minimizers, and our experiments support a commonly held view that this is due to the inherent noise in the gradient estimation. We discuss several strategies to attempt to help large-batch methods eliminate this generalization gap. <|reference_end|>", "<|reference_start|> Opening the Black Box of Deep Neural Networks via Information: Despite their great success, there is still no comprehensive theoretical understanding of learning with Deep Neural Networks (DNNs) or their inner organization. Previous work proposed to analyze DNNs in the \\textit{Information Plane}; i.e., the plane of the Mutual Information values that each layer preserves on the input and output variables. They suggested that the goal of the network is to optimize the Information Bottleneck (IB) tradeoff between compression and prediction, successively, for each layer. In this work we follow up on this idea and demonstrate the effectiveness of the Information-Plane visualization of DNNs. Our main results are: (i) most of the training epochs in standard DL are spent on {\\emph compression} of the input to efficient representation and not on fitting the training labels. (ii) The representation compression phase begins when the training errors becomes small and the Stochastic Gradient Decent (SGD) epochs change from a fast drift to smaller training error into a stochastic relaxation, or random diffusion, constrained by the training error value. (iii) The converged layers lie on or very close to the Information Bottleneck (IB) theoretical bound, and the maps from the input to any hidden layer and from this hidden layer to the output satisfy the IB self-consistent equations. This generalization through noise mechanism is unique to Deep Neural Networks and absent in one layer networks. (iv) The training time is dramatically reduced when adding more hidden layers. Thus the main advantage of the hidden layers is computational. This can be explained by the reduced relaxation time, as this it scales super-linearly (exponentially for simple diffusion) with the information compression from the previous layer. <|reference_end|>", "<|reference_start|> On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length: Stochastic Gradient Descent (SGD) based training of neural networks with a large learning rate or a small batch-size typically ends in well-generalizing, flat regions of the weight space, as indicated by small eigenvalues of the Hessian of the training loss. However, the curvature along the SGD trajectory is poorly understood. An empirical investigation shows that initially SGD visits increasingly sharp regions, reaching a maximum sharpness determined by both the learning rate and the batch-size of SGD. When studying the SGD dynamics in relation to the sharpest directions in this initial phase, we find that the SGD step is large compared to the curvature and commonly fails to minimize the loss along the sharpest directions. Furthermore, using a reduced learning rate along these directions can improve training speed while leading to both sharper and better generalizing solutions compared to vanilla SGD. In summary, our analysis of the dynamics of SGD in the subspace of the sharpest directions shows that they influence the regions that SGD steers to (where larger learning rate or smaller batch size result in wider regions visited), the overall training speed, and the generalization ability of the final model. <|reference_end|>" ]

[ 4, 7, 15, 16 ]

[ "<|reference_start|> A model of computation for mapreduce: In recent years the MapReduce framework has emerged as one of the most widely used parallel computing platforms for processing data on terabyte and petabyte scales. Used daily at companies such as Yahoo!, Google, Amazon, and Facebook, and adopted more recently by several universities, it allows for easy parallelization of data intensive computations over many machines. One key feature of MapReduce that differentiates it from previous models of parallel computation is that it interleaves sequential and parallel computation. We propose a model of efficient computation using the MapReduce paradigm. Since MapReduce is designed for computations over massive data sets, our model limits the number of machines and the memory per machine to be substantially sublinear in the size of the input. On the other hand, we place very loose restrictions on the computational power of of any individual machine---our model allows each machine to perform sequential computations in time polynomial in the size of the original input.\n We compare MapReduce to the PRAM model of computation. We prove a simulation lemma showing that a large class of PRAM algorithms can be efficiently simulated via MapReduce. The strength of MapReduce, however, lies in the fact that it uses both sequential and parallel computation. We demonstrate how algorithms can take advantage of this fact to compute an MST of a dense graph in only two rounds, as opposed to Ω(log(n)) rounds needed in the standard PRAM model. We show how to evaluate a wide class of functions using the MapReduce framework. We conclude by applying this result to show how to compute some basic algorithmic problems such as undirected s-t connectivity in the MapReduce framework. <|reference_end|>", "<|reference_start|> Set cover algorithms for very large datasets: The problem of Set Cover - to find the smallest subcollection of sets that covers some universe - is at the heart of many data and analysis tasks. It arises in a wide range of settings, including operations research, machine learning, planning, data quality and data mining. Although finding an optimal solution is NP-hard, the greedy algorithm is widely used, and typically finds solutions that are close to optimal. However, a direct implementation of the greedy approach, which picks the set with the largest number of uncovered items at each step, does not behave well when the input is very large and disk resident. The greedy algorithm must make many random accesses to disk, which are unpredictable and costly in comparison to linear scans. In order to scale Set Cover to large datasets, we provide a new algorithm which finds a solution that is provably close to that of greedy, but which is much more efficient to implement using modern disk technology. Our experiments show a ten-fold improvement in speed on moderately-sized datasets, and an even greater improvement on larger datasets. <|reference_end|>", "<|reference_start|> Parallel and I/O efficient set covering algorithms: This paper presents the design, analysis, and implementation of parallel and sequential I/O-efficient algorithms for set cover, tying together the line of work on parallel set cover and the line of work on efficient set cover algorithms for large, disk-resident instances.\n Our contributions are twofold: First, we design and analyze a parallel cache-oblivious set-cover algorithm that offers essentially the same approximation guarantees as the standard greedy algorithm, which has the optimal approximation. Our algorithm is the first efficient external-memory or cache-oblivious algorithm for when neither the sets nor the elements fit in memory, leading to I/O cost (cache complexity) equivalent to sorting in the Cache Oblivious or Parallel Cache Oblivious models. The algorithm also implies elow cache misses on parallel hierarchical memories (again, equivalent to sorting). Second, building on this theory, we engineer variants of the theoretical algorithm optimized for different hardware setups. We provide experimental evaluation showing substantial speedups over existing algorithms without compromising the solution's quality. <|reference_end|>", "<|reference_start|> Composable core-sets for Diversity and Coverage Maximization: In this paper we consider efficient construction of \"composable core-sets\" for basic diversity and coverage maximization problems. A core-set for a point-set in a metric space is a subset of the point-set with the property that an approximate solution to the whole point-set can be obtained given the core-set alone. A composable core-set has the property that for a collection of sets, the approximate solution to the union of the sets in the collection can be obtained given the union of the composable core-sets for the point sets in the collection. Using composable core-sets one can obtain efficient solutions to a wide variety of massive data processing applications, including nearest neighbor search, streaming algorithms and map-reduce computation. Our main results are algorithms for constructing composable core-sets for several notions of \"diversity objective functions\", a topic that attracted a significant amount of research over the last few years. The composable core-sets we construct are small and accurate: their approximation factor almost matches that of the best \"off-line\" algorithms for the relevant optimization problems (up to a constant factor). Moreover, we also show applications of our results to diverse nearest neighbor search, streaming algorithms and map-reduce computation. Finally, we show that for an alternative notion of diversity maximization based on the maximum coverage problem small composable core-sets do not exist. <|reference_end|>" ]

[ 11, 23, 25, 27 ]

[ "<|reference_start|> Dynamic Key-Value Memory Networks for Knowledge Tracing: Knowledge Tracing (KT) is a task of tracing evolving knowledge state of students with respect to one or more concepts as they engage in a sequence of learning activities. One important purpose of KT is to personalize the practice sequence to help students learn knowledge concepts efficiently. However, existing methods such as Bayesian Knowledge Tracing and Deep Knowledge Tracing either model knowledge state for each predefined concept separately or fail to pinpoint exactly which concepts a student is good at or unfamiliar with. To solve these problems, this work introduces a new model called Dynamic Key-Value Memory Networks (DKVMN) that can exploit the relationships between underlying concepts and directly output a student's mastery level of each concept. Unlike standard memory-augmented neural networks that facilitate a single memory matrix or two static memory matrices, our model has one static matrix called key, which stores the knowledge concepts and the other dynamic matrix called value, which stores and updates the mastery levels of corresponding concepts. Experiments show that our model consistently outperforms the state-of-the-art model in a range of KT datasets. Moreover, the DKVMN model can automatically discover underlying concepts of exercises typically performed by human annotations and depict the changing knowledge state of a student. <|reference_end|>", "<|reference_start|> One-shot Learning with Memory-Augmented Neural Networks: Despite recent breakthroughs in the applications of deep neural networks, one setting that presents a persistent challenge is that of \"one-shot learning.\" Traditional gradient-based networks require a lot of data to learn, often through extensive iterative training. When new data is encountered, the models must inefficiently relearn their parameters to adequately incorporate the new information without catastrophic interference. Architectures with augmented memory capacities, such as Neural Turing Machines (NTMs), offer the ability to quickly encode and retrieve new information, and hence can potentially obviate the downsides of conventional models. Here, we demonstrate the ability of a memory-augmented neural network to rapidly assimilate new data, and leverage this data to make accurate predictions after only a few samples. We also introduce a new method for accessing an external memory that focuses on memory content, unlike previous methods that additionally use memory location-based focusing mechanisms. <|reference_end|>", "<|reference_start|> Self-Attentive Sequential Recommendation: Sequential dynamics are a key feature of many modern recommender systems, which seek to capture the `context' of users' activities on the basis of actions they have performed recently. To capture such patterns, two approaches have proliferated: Markov Chains (MCs) and Recurrent Neural Networks (RNNs). Markov Chains assume that a user's next action can be predicted on the basis of just their last (or last few) actions, while RNNs in principle allow for longer-term semantics to be uncovered. Generally speaking, MC-based methods perform best in extremely sparse datasets, where model parsimony is critical, while RNNs perform better in denser datasets where higher model complexity is affordable. The goal of our work is to balance these two goals, by proposing a self-attention based sequential model (SASRec) that allows us to capture long-term semantics (like an RNN), but, using an attention mechanism, makes its predictions based on relatively few actions (like an MC). At each time step, SASRec seeks to identify which items are `relevant' from a user's action history, and use them to predict the next item. Extensive empirical studies show that our method outperforms various state-of-the-art sequential models (including MC/CNN/RNN-based approaches) on both sparse and dense datasets. Moreover, the model is an order of magnitude more efficient than comparable CNN/RNN-based models. Visualizations on attention weights also show how our model adaptively handles datasets with various density, and uncovers meaningful patterns in activity sequences. <|reference_end|>", "<|reference_start|> Attention Is All You Need: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data. <|reference_end|>" ]

[ 3, 4, 6, 7 ]

[ "<|reference_start|> Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Images: <|reference_end|>", "<|reference_start|> Attention Guided Anomaly Localization in Images: Anomaly localization is an important problem in computer vision which involves localizing anomalous regions within images with applications in industrial inspection, surveillance, and medical imaging. This task is challenging due to the small sample size and pixel coverage of the anomaly in real-world scenarios. Most prior works need to use anomalous training images to compute a class-specific threshold to localize anomalies. Without the need of anomalous training images, we propose Convolutional Adversarial Variational autoencoder with Guided Attention (CAVGA), which localizes the anomaly with a convolutional latent variable to preserve the spatial information. In the unsupervised setting, we propose an attention expansion loss where we encourage CAVGA to focus on all normal regions in the image. Furthermore, in the weakly-supervised setting we propose a complementary guided attention loss, where we encourage the attention map to focus on all normal regions while minimizing the attention map corresponding to anomalous regions in the image. CAVGA outperforms the state-of-the-art (SOTA) anomaly localization methods on MVTec Anomaly Detection (MVTAD), modified ShanghaiTech Campus (mSTC) and Large-scale Attention based Glaucoma (LAG) datasets in the unsupervised setting and when using only 2% anomalous images in the weakly-supervised setting. CAVGA also outperforms SOTA anomaly detection methods on the MNIST, CIFAR-10, Fashion-MNIST, MVTAD, mSTC and LAG datasets. <|reference_end|>", "<|reference_start|> FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows: Unsupervised anomaly detection and localization is crucial to the practical application when collecting and labeling sufficient anomaly data is infeasible. Most existing representation-based approaches extract normal image features with a deep convolutional neural network and characterize the corresponding distribution through non-parametric distribution estimation methods. The anomaly score is calculated by measuring the distance between the feature of the test image and the estimated distribution. However, current methods can not effectively map image features to a tractable base distribution and ignore the relationship between local and global features which are important to identify anomalies. To this end, we propose FastFlow implemented with 2D normalizing flows and use it as the probability distribution estimator. Our FastFlow can be used as a plug-in module with arbitrary deep feature extractors such as ResNet and vision transformer for unsupervised anomaly detection and localization. In training phase, FastFlow learns to transform the input visual feature into a tractable distribution and obtains the likelihood to recognize anomalies in inference phase. Extensive experimental results on the MVTec AD dataset show that FastFlow surpasses previous state-of-the-art methods in terms of accuracy and inference efficiency with various backbone networks. Our approach achieves 99.4% AUC in anomaly detection with high inference efficiency. <|reference_end|>", "<|reference_start|> {Denoising autoencoders for unsupervised anomaly detection in brain MRI: Pathological brain lesions exhibit diverse appearance in brain images, making it difficult to train supervised detection solutions due to the lack of comprehensive data and annotations. Thus, in this work we tackle unsupervised anomaly detection, using only healthy data for training with the aim of detecting unseen anomalies at test time. Many current approaches employ autoencoders with restrictive architectures (i.e. containing information bottlenecks) that tend to give poor reconstructions of not only the anomalous but also the normal parts of the brain. Instead, we investigate classical denoising autoencoder models that do not require bottlenecks and can employ skip connections to give high resolution fidelity. We design a simple noise generation method of upscaling low-resolution noise that enables high-quality reconstructions. We find that with appropriate noise generation, denoising autoencoder reconstruction errors generalize to hyperintense lesion segmentation and reach state of the art performance for unsupervised tumor detection in brain MRI data, beating more complex methods such as variational autoencoders. We believe this provides a strong and easy-to-implement baseline for further research into unsupervised anomaly detection. <|reference_end|>" ]

[ 0, 27, 48, 63 ]

[ "<|reference_start|> Sublinear-Time Decremental Algorithms for Single-Source Reachability and Shortest Paths on Directed Graphs: We consider dynamic algorithms for maintaining Single-Source Reachability (SSR) and approximate Single-Source Shortest Paths (SSSP) on $n$-node $m$-edge directed graphs under edge deletions (decremental algorithms). The previous fastest algorithm for SSR and SSSP goes back three decades to Even and Shiloach [JACM 1981]; it has $ O(1) $ query time and $ O (mn) $ total update time (i.e., linear amortized update time if all edges are deleted). This algorithm serves as a building block for several other dynamic algorithms. The question whether its total update time can be improved is a major, long standing, open problem. In this paper, we answer this question affirmatively. We obtain a randomized algorithm with an expected total update time of $ O(\\min (m^{7/6} n^{2/3 + o(1)}, m^{3/4} n^{5/4 + o(1)}) ) = O (m n^{9/10 + o(1)}) $ for SSR and $(1+\\epsilon)$-approximate SSSP if the edge weights are integers from $ 1 $ to $ W \\leq 2^{\\log^c{n}} $ and $ \\epsilon \\geq 1 / \\log^c{n} $ for some constant $ c $. We also extend our algorithm to achieve roughly the same running time for Strongly Connected Components (SCC), improving the algorithm of Roditty and Zwick [FOCS 2002]. Our algorithm is most efficient for sparse and dense graphs. When $ m = \\Theta(n) $ its running time is $ O (n^{1 + 5/6 + o(1)}) $ and when $ m = \\Theta(n^2) $ its running time is $ O (n^{2 + 3/4 + o(1)}) $. For SSR we also obtain an algorithm that is faster for dense graphs and has a total update time of $ O ( m^{2/3} n^{4/3 + o(1)} + m^{3/7} n^{12/7 + o(1)}) $ which is $ O (n^{2 + 2/3}) $ when $ m = \\Theta(n^2) $. All our algorithms have constant query time in the worst case and are correct with high probability against an oblivious adversary. <|reference_end|>", "<|reference_start|> Parallel Reachability in Almost Linear Work and Square Root Depth: In this paper we provide a parallel algorithm that given any $n$-node $m$-edge directed graph and source vertex $s$ computes all vertices reachable from $s$ with $\\tilde{O}(m)$ work and $n^{1/2 + o(1)}$ depth with high probability in $n$ . This algorithm also computes a set of $\\tilde{O}(n)$ edges which when added to the graph preserves reachability and ensures that the diameter of the resulting graph is at most $n^{1/2 + o(1)}$. Our result improves upon the previous best known almost linear work reachability algorithm due to Fineman which had depth $\\tilde{O}(n^{2/3})$. Further, we show how to leverage this algorithm to achieve improved distributed algorithms for single source reachability in the CONGEST model. In particular, we provide a distributed algorithm that given a $n$-node digraph of undirected hop-diameter $D$ solves the single source reachability problem with $\\tilde{O}(n^{1/2} + n^{1/3 + o(1)} D^{2/3})$ rounds of the communication in the CONGEST model with high probability in $n$. Our algorithm is nearly optimal whenever $D = O(n^{1/4 - \\epsilon})$ for any constant $\\epsilon > 0$ and is the first nearly optimal algorithm for general graphs whose diameter is $\\Omega(n^\\delta)$ for any constant $\\delta$. <|reference_end|>", "<|reference_start|> Better Lower Bounds for Shortcut Sets and Additive Spanners via an Improved Alternation Product: We obtain improved lower bounds for additive spanners, additive emulators, and diameter-reducing shortcut sets. Spanners and emulators are sparse graphs that approximately preserve the distances of a given graph. A shortcut set is a set of edges that when added to a directed graph, decreases its diameter. The previous best known lower bounds for these three structures are given by Huang and Pettie [SWAT 2018]. For $O(n)$-sized spanners, we improve the lower bound on the additive stretch from $\\Omega(n^{1/11})$ to $\\Omega(n^{2/21})$. For $O(n)$-sized emulators, we improve the lower bound on the additive stretch from $\\Omega(n^{1/18})$ to $\\Omega(n^{1/16})$. For $O(m)$-sized shortcut sets, we improve the lower bound on the graph diameter from $\\Omega(n^{1/11})$ to $\\Omega(n^{1/8})$. Our key technical contribution, which is the basis of all of our bounds, is an improvement of a graph product known as an alternation product. <|reference_end|>", "<|reference_start|> Having Hope in Hops: New Spanners, Preservers and Lower Bounds for Hopsets: Hopsets and spanners are fundamental graph structures, playing a key role in shortest path computation, distributed communication, and more. A (near-exact) hopset for a given graph $G$ is a (small) subset of weighted edges $H$ that when added to the graph $G$ reduces the number of hops (edges) of near-exact shortest paths. Spanners and distance preservers, on the other hand, ask for removing many edges from the graph while approximately preserving shortest path distances. We provide a general reduction scheme from graph hopsets to the known metric compression schemes of spanners, emulators and distance preservers. Consequently, we get new and improved upper bound constructions for the latter, as well as, new lower bound results for hopsets. Our work makes a significant progress on the tantalizing open problem concerning the formal connection between hopsets and spanners, e.g., as posed by Elkin and Neiman [Bull. EATCS 2020]. <|reference_end|>" ]

[ 2, 5, 23, 30 ]

[ "<|reference_start|> {Fast Bellman Updates for Robust MDPs: We describe two efficient, and exact, algorithms for computing Bellman updates in robust Markov decision processes (MDPs). The first algorithm uses a homotopy continuation method to compute updates for L1-constrained s, a-rectangular ambiguity sets. It runs in quasi-linear time for plain L1 norms and also generalizes to weighted L1 norms. The second algorithm uses bisection to compute updates for robust MDPs with s-rectangular ambiguity sets. This algorithm, when combined with the homotopy method, also has a quasi-linear runtime. Unlike previous methods, our algorithms compute the primal solution in addition to the optimal objective value, which makes them useful in policy iteration methods. Our experimental results indicate that the proposed methods are over 1,000 times faster than Gurobi, a state-of-the-art commercial optimization package, for small instances, and the performance gap grows considerably with problem size. <|reference_end|>", "<|reference_start|> A Model Selection Approach for Corruption Robust Reinforcement Learning: We develop a model selection approach to tackle reinforcement learning with adversarial corruption in both transition and reward. For finite-horizon tabular MDPs, without prior knowledge on the total amount of corruption, our algorithm achieves a regret bound of $\\widetilde{\\mathcal{O}}(\\min\\{\\frac{1}{\\Delta}, \\sqrt{T}\\}+C)$ where $T$ is the number of episodes, $C$ is the total amount of corruption, and $\\Delta$ is the reward gap between the best and the second-best policy. This is the first worst-case optimal bound achieved without knowledge of $C$, improving previous results of Lykouris et al. (2021); Chen et al. (2021); Wu et al. (2021). For finite-horizon linear MDPs, we develop a computationally efficient algorithm with a regret bound of $\\widetilde{\\mathcal{O}}(\\sqrt{(1+C)T})$, and another computationally inefficient one with $\\widetilde{\\mathcal{O}}(\\sqrt{T}+C)$, improving the result of Lykouris et al. (2021) and answering an open question by Zhang et al. (2021b). Finally, our model selection framework can be easily applied to other settings including linear bandits, linear contextual bandits, and MDPs with general function approximation, leading to several improved or new results. <|reference_end|>", "<|reference_start|> Double Pessimism is Provably Efficient for Distributionally Robust Offline Reinforcement Learning: Generic Algorithm and Robust Partial Coverage: In this paper, we study distributionally robust offline reinforcement learning (robust offline RL), which seeks to find an optimal policy purely from an offline dataset that can perform well in perturbed environments. In specific, we propose a generic algorithm framework called Doubly Pessimistic Model-based Policy Optimization ($P^2MPO$), which features a novel combination of a flexible model estimation subroutine and a doubly pessimistic policy optimization step. Notably, the double pessimism principle is crucial to overcome the distributional shifts incurred by (i) the mismatch between the behavior policy and the target policies; and (ii) the perturbation of the nominal model. Under certain accuracy conditions on the model estimation subroutine, we prove that $P^2MPO$ is sample-efficient with robust partial coverage data, which only requires the offline data to have good coverage of the distributions induced by the optimal robust policy and the perturbed models around the nominal model. By tailoring specific model estimation subroutines for concrete examples of RMDPs, including tabular RMDPs, factored RMDPs, kernel and neural RMDPs, we prove that $P^2MPO$ enjoys a $\\tilde{\\mathcal{O}}(n^{-1/2})$ convergence rate, where $n$ is the dataset size. We highlight that all these examples, except tabular RMDPs, are first identified and proven tractable by this work. Furthermore, we continue our study of robust offline RL in the robust Markov games (RMGs). By extending the double pessimism principle identified for single-agent RMDPs, we propose another algorithm framework that can efficiently find the robust Nash equilibria among players using only robust unilateral (partial) coverage data. To our best knowledge, this work proposes the first general learning principle -- double pessimism -- for robust offline RL and shows that it is provably efficient with general function approximation. <|reference_end|>", "<|reference_start|> Almost Optimal Algorithms for Linear Stochastic Bandits with Heavy-Tailed Payoffs: In linear stochastic bandits, it is commonly assumed that payoffs are with sub-Gaussian noises. In this paper, under a weaker assumption on noises, we study the problem of \\underline{lin}ear stochastic {\\underline b}andits with h{\\underline e}avy-{\\underline t}ailed payoffs (LinBET), where the distributions have finite moments of order $1+\\epsilon$, for some $\\epsilon\\in (0,1]$. We rigorously analyze the regret lower bound of LinBET as $\\Omega(T^{\\frac{1}{1+\\epsilon}})$, implying that finite moments of order 2 (i.e., finite variances) yield the bound of $\\Omega(\\sqrt{T})$, with $T$ being the total number of rounds to play bandits. The provided lower bound also indicates that the state-of-the-art algorithms for LinBET are far from optimal. By adopting median of means with a well-designed allocation of decisions and truncation based on historical information, we develop two novel bandit algorithms, where the regret upper bounds match the lower bound up to polylogarithmic factors. To the best of our knowledge, we are the first to solve LinBET optimally in the sense of the polynomial order on $T$. Our proposed algorithms are evaluated based on synthetic datasets, and outperform the state-of-the-art results. <|reference_end|>" ]

[ 7, 19, 26, 39 ]

[ "<|reference_start|> Structuring contemporary remote sensing image fusion: The exploitation of multi-sensor images at pixel level is a widely implemented research field in Earth observation. In this context, image fusion plays an important role since it effectively combines complementary image content to enhance information contained in the individual datasets. This article presents an overview of the existing fusion techniques and their achievements for Earth scientists. This research started off with the compilation of a database on remote sensing image fusion journal publications. Research results were exploited, grouping the literature into different aspects of relevance. Six categories of information have been built according to the journal, the application, sensors that provided the images used in the case study, applied fusion techniques, areas of achievement, and on-going research highlighting unresolved questions and current science. This resulted in an overview on the categorisation of image fusion techniques, explanation of the various approaches used within a certain category, and description of particularities when dealing with the fusion of optical and radar imagery. Even though many researchers intend to find the best algorithm, there is a greater need to define an appropriate workflow prior to processing the imagery with the knowledge in all related fields, that is, remote sensing image fusion and the desired application to address the different aspects of error propagation. <|reference_end|>", "<|reference_start|> A new look at IHS-like image fusion methods: <|reference_end|>", "<|reference_start|> Fast and Efficient Panchromatic Sharpening: Certain sensors such as IKONOS produce panchromatic and multispectral (MS) images at different spatial resolutions. Several efforts have been made to increase the resolution of these MS images using panchromatic image information. In this paper, we present a fast and efficient panchromatic sharpening method that accurately estimates missing high-frequency components. We also use a postprocessing technique to correct color distortion. Experimental results show that the proposed method produced high-quality images and outperformed existing panchromatic sharpening methods in terms of objective quality measures such as universal image quality index, Q4, relative average spectral error, Erreur Relative Globale Adimensionnelle de SynthE¿se, and correlation. <|reference_end|>", "<|reference_start|> {Fusion of high spatial and spectral resolution images: The ARSIS concept and its implementation: In various applications of remote sensing, when high spatial resolution is required in addition with classification results, sensorfusion is a solution. From a set of images with different spatial and spectral resolutions, the aim is to synthesize images with the highest spatial resolution available in the set and with an appropriate spectral content. Several sensor fusion methods exist; most of them improve the spatial resolution but provide poor quality of the spectral content of the resulting image. Based on a multiresolution modeling of the information, the ARsIs concept [from its French name \"Am6lioration de la R6solution Spatiale par Injection de Structures\") was designed with the aim of improving the spatial resolution together with a high quality in the spectral content ofthe synthesized images. The general case for the application of this concept is described. A quantitative comparison of all presented methods is achieved for a SPOT image. Another example of the fusion of SPOTXS (20-m) and KVR-1000 (2-m) images is given. Practical information for the implementation of the wavelet transform, the multiresolution analysis, and the ARSIS concept by practitioners is given with particular relevance to SPOT and Landsat imagew. lntroductlon <|reference_end|>" ]

[ 1, 4, 15, 36 ]

[ "<|reference_start|> Control of variable-speed wind turbines: standard and adaptive techniques for maximizing energy capture: This article considers an adaptive control scheme previously developed for region 2 control of a variable speed wind turbine. In this paper, the question of theoretical stability of the torque controller is addressed, showing that the rotor speed is asymptotically stable under the torque control law in the constant wind speed input case and L/sub 2/ stable with respect to time-varying wind input. Further, a method is derived for selecting /spl gamma//sub /spl Delta/M/ in the gain adaptation law to guarantee convergence of the adaptive gain M to its optimal value M*. <|reference_end|>", "<|reference_start|> Proceedings of the American control conference (ACC): <|reference_end|>", "<|reference_start|> {Real-time optimization by extremum-seeking control: From the Publisher: \nExtremum seeking is a method of adaptive control outside the classical paradigm or model reference and related schemes. The unique feature of extremum seeking as an optimization tool is that is is an on-line methodology, performed by feedback, and with the ability to achieve rapid convergence. A second distinction between classical adaptive control and extremum seeking is that the latter is not model based. Its non-model based character explains the remarkable popularity of extremum seeking in the last half decade: the recent applications in fluid flow combustion, and biomedical systems are all characterized by complex, unreliable models. Written by authorities in the field and pioneers in adaptive nonlinear control systems, this book presents both significant theoretic value and important practical potential. <|reference_end|>", "<|reference_start|> Evaluation of log‐of‐power extremum seeking control for wind turbines using large eddy simulations: <|reference_end|>" ]

[ 1, 2, 4, 10 ]

[ "<|reference_start|> Audio-visual biometrics: Biometric characteristics can be utilized in order to enable reliable and robust-to-impostor-attacks person recognition. Speaker recognition technology is commonly utilized in various systems enabling natural human computer interaction. The majority of the speaker recognition systems rely only on acoustic information, ignoring the visual modality. However, visual information conveys correlated and complimentary information to the audio information and its integration into a recognition system can potentially increase the system's performance, especially in the presence of adverse acoustic conditions. Acoustic and visual biometric signals, such as the person's voice and face, can be obtained using unobtrusive and user-friendly procedures and low-cost sensors. Developing unobtrusive biometric systems makes biometric technology more socially acceptable and accelerates its integration into every day life. In this paper, we describe the main components of audio-visual biometric systems, review existing systems and their performance, and discuss future research and development directions in this area <|reference_end|>", "<|reference_start|> Improved Speaker Independent Lip Reading Using Speaker Adaptive Training and Deep Neural Networks: Recent improvements in tracking and feature extraction mean that speaker-dependent lip-reading of continuous speech using a medium size vocabulary (around 1000 words) is realistic. However, the recognition of previously unseen speakers has been found to be a very challenging task, because of the large variation in lip-shapes across speakers and the lack of large, tracked databases of visual features, which are very expensive to produce. By adapting a technique that is established in speech recognition but has not previously been used in lip-reading, we show that error-rates for speaker-independent lip-reading can be very significantly reduced. Furthermore, we show that error-rates can be even further reduced by the additional use of Deep Neural Networks (DNN). We also find that there is no need to map phonemes to visemes for context-dependent visual speech transcription. <|reference_end|>", "<|reference_start|> Speaker identification by lipreading: This paper describes a new approach for speaker identification based on lipreading. Visual features are extracted from image sequences of the talking face and consist of shape parameters which describe the lip boundary and intensity parameters which describe the grey-level distribution of the mouth area. Intensity information is based on principal component analysis using eigenspaces which deform with the shape model. The extracted parameters account for both, speech dependent and speaker dependent information. We built spatio-temporal speaker models based on these features, using HMMs with mixtures of Gaussians. Promising results were obtained for text dependent and text independent speaker identification tests performed on a small video database. <|reference_end|>", "<|reference_start|> Lipreading with DenseNet and resBi-LSTM: <|reference_end|>" ]

[ 0, 14, 21, 24 ]

[ "<|reference_start|> 2011 IEEE Vehicular Networking Conference, IEEE VNC 2011, Amsterdam, The Netherlands, November 14-16, 2011: <|reference_end|>", "<|reference_start|> On the Performance of Secure Vehicular Communication Systems: Vehicular communication (VC) systems are being developed primarily to enhance transportation safety and efficiency. Vehicle-to-vehicle communication, in particular, frequent cooperative awareness messages or safety beacons, has been considered over the past years as a main approach. Meanwhile, the need to provide security and to safeguard users' privacy is well understood, and security architectures for VC systems have been proposed. Although technical approaches to secure VC have several commonalities and a consensus has formed, there are critical questions that have remained largely unanswered: Are the proposed security and privacy schemes practical? Can the secured VC systems support the VC-enabled applications as effectively as unsecured VC would? How should security be designed so that its integration into a VC system has a limited effect on the system's performance? In this paper, we provide answers to these questions, investigating the joint effect of a set of system parameters and components. We consider the state-of-the-art approach in secure VC, and we evaluate analytically and through simulations the interdependencies among components and system characteristics. Overall, we identify key design choices for the deployment of efficient, effective, and secure VC systems. <|reference_end|>", "<|reference_start|> Wireless channel-based message authentication: Inter-vehicle communication has attracted a lot of attention in the past. A major concern is the security and especially the integrity and authenticity of messages. Current standards and proposals in literature leverage asymmetric cryptographic mechanisms to achieve this, which is costly both in terms of consumed computational power, bandwidth, and introduced delay. We present a novel idea to use physical characteristics of the wireless channel to verify subsequent messages after initial trust in a first message has been established cryptographically. In this paper, we sketch the concept and provide a first evaluation on its potential for saving named resources. <|reference_end|>", "<|reference_start|> Multi-user broadcast authentication in wireless sensor networks: Broadcast authentication is a critical security service in wireless sensor networks (WSNs), as it allows the mobile users of WSNs to broadcast messages to multiple sensor nodes in a secure way. Although symmetric-key- based solutions such as muTESLA and multilevel muTESLA have been proposed, they all suffer from severe energy- depletion attacks resulted from the nature of delayed message authentication. This paper presents several efficient public-key-based schemes to achieve immediate broadcast authentication and thus avoid the security vulnerability intrinsic to muTESLA-like schemes. Our schemes are built upon the unique integration of several cryptographic techniques, including the Bloom filter, the partial message recovery signature scheme and the Merkle hash tree. We prove the effectiveness and efficiency of the proposed schemes by a comprehensive quantitative analysis of their energy consumption in both computation and communication. <|reference_end|>" ]

[ 1, 2, 4, 7 ]

[ "<|reference_start|> VQA: Visual Question Answering: We propose the task of free-form and open-ended Visual Question Answering (VQA). Given an image and a natural language question about the image, the task is to provide an accurate natural language answer. Mirroring real-world scenarios, such as helping the visually impaired, both the questions and answers are open-ended. Visual questions selectively target different areas of an image, including background details and underlying context. As a result, a system that succeeds at VQA typically needs a more detailed understanding of the image and complex reasoning than a system producing generic image captions. Moreover, VQA is amenable to automatic evaluation, since many open-ended answers contain only a few words or a closed set of answers that can be provided in a multiple-choice format. We provide a dataset containing ~0.25M images, ~0.76M questions, and ~10M answers (www.visualqa.org), and discuss the information it provides. Numerous baselines and methods for VQA are provided and compared with human performance. Our VQA demo is available on CloudCV (http://cloudcv.org/vqa). <|reference_end|>", "<|reference_start|> Prioritized Experience Replay: Experience replay lets online reinforcement learning agents remember and reuse experiences from the past. In prior work, experience transitions were uniformly sampled from a replay memory. However, this approach simply replays transitions at the same frequency that they were originally experienced, regardless of their significance. In this paper we develop a framework for prioritizing experience, so as to replay important transitions more frequently, and therefore learn more efficiently. We use prioritized experience replay in Deep Q-Networks (DQN), a reinforcement learning algorithm that achieved human-level performance across many Atari games. DQN with prioritized experience replay achieves a new state-of-the-art, outperforming DQN with uniform replay on 41 out of 49 games. <|reference_end|>", "<|reference_start|> Hierarchical Reinforcement Learning of Dialogue Policies in a development environment for dialogue systems: REALL-DUDE: A pitcher's mound on, which a pitcher in the sport of Baseball stands while pitching the ball, is collapsible, and is made up of several wedge-shaped segments which are connected to a central hub in an easily releasable manner. The segments and the central hub are quite rigid yet are of a size and weight which makes them reasonably portable. The surface of the pitcher's mound can be such as to absorb impact when struck by a ball and to provide good footing for the pitcher. <|reference_end|>", "<|reference_start|> Evaluation of a hierarchical reinforcement learning spoken dialogue system: <|reference_end|>" ]

[ 2, 25, 29, 30 ]

[ "<|reference_start|> Far-field automatic speech recognition: The machine recognition of speech spoken at a distance from the microphones, known as far-field automatic speech recognition (ASR), has received a significant increase in attention in science and industry, which caused or was caused by an equally significant improvement in recognition accuracy. Meanwhile, it has entered the consumer market with digital home assistants with a spoken language interface being its most prominent application. Speech recorded at a distance is affected by various acoustic distortions, and consequently, quite different processing pipelines have emerged compared with ASR for close-talk speech. A signal enhancement front end for dereverberation, source separation, and acoustic beamforming is employed to clean up the speech, and the back-end ASR engine is robustified by multicondition training and adaptation. We will also describe the so-called end-to-end approach to ASR, which is a new promising architecture that has recently been extended to the far-field scenario. This tutorial article gives an account of the algorithms used to enable accurate speech recognition from a distance, and it will be seen that, although deep learning has a significant share in the technological breakthroughs, a clever combination with traditional signal processing can lead to surprisingly effective solutions. <|reference_end|>", "<|reference_start|> Training Data Generation with DOA-based Selecting and Remixing for Unsupervised Training of Deep Separation Models: <|reference_end|>", "<|reference_start|> Neural full-rank spatial covariance analysis for blind source separation: This paper describes aneural blind source separation (BSS) method based on amortized variational inference (AVI) of a non-linear generative model of mixture signals. A classical statistical approach to BSS is to fit a linear generative model that consists of spatial and source models representing the inter-channel covariances and power spectral densities of sources, respectively. Although the variational autoencoder (VAE) has successfully been used as a non-linear source model with latent features, it should be pretrained from a sufficient amount of isolated signals. Our method, in contrast, enables the VAE-based source model to be trained only from mixture signals. Specifically, we introduce a neural mixture-to-feature inference model that directly infers the latent features from the observed mixture and integrate it with a neural feature-to-mixture generative model consisting of a full-rank spatial model and a VAE-based source model. All the models are optimized jointly such that the likelihood for the training mixtures is maximized in the framework of AVI. Once the inference model is optimized, it can be used for estimating the latent features of sources included in unseen mixture signals. The experimental results show that the proposed method outperformed the state-of-the-art BSS methods based on linear generative models and was comparable to a method based on supervised learning of the VAE-based sourcemodel. <|reference_end|>", "<|reference_start|> Bootstrapping single-channel source separation via unsupervised spatial clustering on stereo mixtures: Separating an audio scene into isolated sources is a fundamental problem in computer audition, analogous to image segmentation in visual scene analysis. Source separation systems based on deep learning are currently the most successful approaches for solving the underdetermined separation problem, where there are more sources than channels. Traditionally, such systems are trained on sound mixtures where the ground truth decomposition is already known. Since most real-world recordings do not have such a decomposition available, this limits the range of mixtures one can train on, and the range of mixtures the learned models may successfully separate. In this work, we use a simple blind spatial source separation algorithm to generate estimated decompositions of stereo mixtures. These estimates, together with a weighting scheme in the time-frequency domain, based on confidence in the separation quality, are used to train a deep learning model that can be used for single-channel separation, where no source direction information is available. This demonstrates how a simple cue such as the direction of origin of source can be used to bootstrap a model for source separation that can be used in situations where that cue is not available. <|reference_end|>" ]

[ 1, 14, 16, 23 ]

[ "<|reference_start|> Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications: <|reference_end|>", "<|reference_start|> Open Issues in Router Buffer Sizing: Recent research results suggest that the buffers of router interfaces can be made very small, much less than the link's and width-delay product, without causing a utilization loss, as long as the link carries many TCP flows. In this letter we raise some concerns about the previous recommendation. We show that the use of such small buffers can lead to excessively high loss rates (up to 5%-15%in our simulations) in congested access links that carry many flows. Even if the link is fully utilized, small buffers lead to lower throughput for most large TCP flows, and significant variability in the per-flow throughput and transfer latency. We also discuss some important issues in router buffer sizing that are often ignored <|reference_end|>", "<|reference_start|> Sizing Router Buffers: All Internet routers contain buffers to hold packets during times of congestion. Today, the size of the buffers is determined by the dynamics of TCP's congestion control algorithm. In particular, the goal is to make sure that when a link is congested, it is busy 100% of the time; which is equivalent to making sure its buffer never goes empty. A widely used rule-of-thumb states that each link needs a buffer of size B = overlineRTT x C, where overlineRTT is the average round-trip time of a flow passing across the link, and C is the data rate of the link. For example, a 10Gb/s router linecard needs approximately 250ms x 10Gb/s = 2.5Gbits of buffers; and the amount of buffering grows linearly with the line-rate. Such large buffers are challenging for router manufacturers, who must use large, slow, off-chip DRAMs. And queueing delays can be long, have high variance, and may destabilize the congestion control algorithms. In this paper we argue that the rule-of-thumb (B = (overlineRTT x C) is now outdated and incorrect for backbone routers. This is because of the large number of flows (TCP connections) multiplexed together on a single backbone link. Using theory, simulation and experiments on a network of real routers, we show that a link with n flows requires no more than B = (overlineRTT x C) √n, for long-lived or short-lived TCP flows. The consequences on router design are enormous: A 2.5Gb/s link carrying 10,000 flows could reduce its buffers by 99% with negligible difference in throughput; and a 10Gb/s link carrying 50,000 flows requires only 10Mbits of buffering, which can easily be implemented using fast, on-chip SRAM. <|reference_end|>", "<|reference_start|> Buffer Sizing for TCP Flows in 802.11e WLANs: We consider the task of sizing buffers for TCP flows in 802.11e WLANs. A number of fundamental new issues arise compared to wired networks. These include that the mean service rate is dependent on the level of channel contention and packet inter-service times vary stochastically due to the random nature of CSMA/CA operation. We find that these considerations lead naturally to a requirement for adaptation of buffer sizes in response to changing network conditions. <|reference_end|>" ]

[ 7, 18, 25, 35 ]

[ "<|reference_start|> Change detection in 3D point clouds acquired by a mobile mapping system: Abstract. Thanks to the development of Mobile mapping systems (MMS), street object recognition, classification, modelling and related studies have become hot topics recently. There has been increasing interest in detecting changes between mobile laser scanning (MLS) point clouds in complex urban areas. A method based on the consistency between the occupancies of space computed from different datasets is proposed. First occupancy of scan rays (empty, occupied, unknown) are defined while considering the accuracy of measurement and registration. Then the occupancy of scan rays are fused using the Weighted Dempster–Shafer theory (WDST). Finally, the consistency between different datasets is obtained by comparing the occupancy at points from one dataset with the fused occupancy of neighbouring rays from the other dataset. Change detection results are compared with a conventional point to triangle (PTT) distance method. Changes at point level are detected fully automatically. The proposed approach allows to detect changes at large scales in urban scenes with fine detail and more importantly, distinguish real changes from occlusions. <|reference_end|>", "<|reference_start|> Image Based Detection of Geometric Changes in Urban Environments: In this paper, we propose an efficient technique to detect changes in the geometry of an urban environment using some images observing its current state. The proposed method can be used to significantly optimize the process of updating the 3D model of a city changing over time, by restricting this process to only those areas where changes are detected. With this application in mind, we designed our algorithm to specifically detect only structural changes in the environment, ignoring any changes in its appearance, and ignoring also all the changes which are not relevant for update purposes, such as cars, people etc. As a by-product, the algorithm also provides a coarse geometry of the detected changes. The performance of the proposed method was tested on four different kinds of urban environments and compared with two alternative techniques. <|reference_end|>", "<|reference_start|> Image Based Detection of Geometric Changes in Urban Environments: In this paper, we propose an efficient technique to detect changes in the geometry of an urban environment using some images observing its current state. The proposed method can be used to significantly optimize the process of updating the 3D model of a city changing over time, by restricting this process to only those areas where changes are detected. With this application in mind, we designed our algorithm to specifically detect only structural changes in the environment, ignoring any changes in its appearance, and ignoring also all the changes which are not relevant for update purposes, such as cars, people etc. As a by-product, the algorithm also provides a coarse geometry of the detected changes. The performance of the proposed method was tested on four different kinds of urban environments and compared with two alternative techniques. <|reference_end|>", "<|reference_start|> Detection of geometric temporal changes in point clouds: Detecting geometric changes between two 3D captures of the same location performed at different moments is a critical operation for all systems requiring a precise segmentation between change and no‐change regions. Such application scenarios include 3D surface reconstruction, environment monitoring, natural events management and forensic science. Unfortunately, typical 3D scanning setups cannot provide any one‐to‐one mapping between measured samples in static regions: in particular, both extrinsic and intrinsic sensor parameters may vary over time while sensor noise and outliers additionally corrupt the data. In this paper, we adopt a multi‐scale approach to robustly tackle these issues. Starting from two point clouds, we first remove outliers using a probabilistic operator. Then, we detect the actual change using the implicit surface defined by the point clouds under a Growing Least Square reconstruction that, compared to the classical proximity measure, offers a more robust change/no‐change characterization near the temporal intersection of the scans and in the areas exhibiting different sampling density and direction. The resulting classification is enhanced with a spatial reasoning step to solve critical geometric configurations that are common in man‐made environments. We validate our approach on a synthetic test case and on a collection of real data sets acquired using commodity hardware. Finally, we show how 3D reconstruction benefits from the resulting precise change/no‐change segmentation. <|reference_end|>" ]

[ 13, 15, 20, 21 ]

[ "<|reference_start|> Generalizing capacity: new definitions and capacity theorems for composite channels: We consider three capacity definitions for composite channels with channel side information at the receiver. A composite channel consists of a collection of different channels with a distribution characterizing the probability that each channel is in operation. The Shannon capacity of a channel is the highest rate asymptotically achievable with arbitrarily small error probability. Under this definition, the transmission strategy used to achieve the capacity must achieve arbitrarily small error probability for all channels in the collection comprising the composite channel. The resulting capacity is dominated by the worst channel in its collection, no matter how unlikely that channel is. We, therefore, broaden the definition of capacity to allow for some outage. The capacity versus outage is the highest rate asymptotically achievable with a given probability of decoder-recognized outage. The expected capacity is the highest average rate asymptotically achievable with a single encoder and multiple decoders, where channel side information determines the channel in use. The expected capacity is a generalization of capacity versus outage since codes designed for capacity versus outage decode at one of two rates (rate zero when the channel is in outage and the target rate otherwise) while codes designed for expected capacity can decode at many rates. Expected capacity equals Shannon capacity for channels governed by a stationary ergodic random process but is typically greater for general channels. The capacity versus outage and expected capacity definitions relax the constraint that all transmitted information must be decoded at the receiver. We derive channel coding theorems for these capacity definitions through information density and provide numerical examples to highlight their connections and differences. We also discuss the implications of these alternative capacity definitions for end-to-end distortion, source-channel coding, and separation. <|reference_end|>", "<|reference_start|> Information theory: Coding theorems for discrete memoryless Systems: Csiszr and Krner's book is widely regarded as a classic in the field of information theory, providing deep insights and expert treatment of the key theoretical issues. It includes in-depth coverage of the mathematics of reliable information transmission, both in two-terminal and multi-terminal network scenarios. Updated and considerably expanded, this new edition presents unique discussions of information theoretic secrecy and of zero-error information theory, including the deep connections of the latter with extremal combinatorics. The presentations of all core subjects are self contained, even the advanced topics, which helps readers to understand the important connections between seemingly different problems. Finally, 320 end-of-chapter problems, together with helpful solving hints, allow readers to develop a full command of the mathematical techniques. It is an ideal resource for graduate students and researchers in electrical and electronic engineering, computer science and applied mathematics. <|reference_end|>", "<|reference_start|> Generalizing the Fano inequality: The Fano inequality gives a lower bound on the mutual information between two random variables that take values on an M-element set, provided at least one of the random variables is equiprobable. The authors show several simple lower bounds on mutual information which do not assume such a restriction. In particular, this ran be accomplished by replacing log M with the infinite-order Renyi entropy in the Fano inequality. Applications to hypothesis testing are exhibited along with bounds on mutual information in terms of the a priori and a posteriori error probabilities. > <|reference_end|>", "<|reference_start|> General Broadcast Channels with Degraded Message Sets: A broadcast channel with one sender and two receivers is considered. Three independent messages are to be transmitted over this channel: one common message which is meant for both receivers, and one private message for each of them. The coding theorem and strong converse for this communication situation is proved for the case when one of the private messages has rate zero. <|reference_end|>" ]

[ 1, 2, 3, 11 ]

[ "<|reference_start|> {The variational formulation of the Fokker--Planck equation: The Fokker--Planck equation, or forward Kolmogorov equation, describes the evolution of the probability density for a stochastic process associated with an Ito stochastic differential equation. It ... <|reference_end|>", "<|reference_start|> Annealed Stein Variational Gradient Descent: Particle based optimization algorithms have recently been developed as sampling methods that iteratively update a set of particles to approximate a target distribution. In particular Stein variational gradient descent has gained attention in the approximate inference literature for its flexibility and accuracy. We empirically explore the ability of this method to sample from multi-modal distributions and focus on two important issues: (i) the inability of the particles to escape from local modes and (ii) the inefficacy in reproducing the density of the different regions. We propose an annealing schedule to solve these issues and show, through various experiments, how this simple solution leads to significant improvements in mode coverage, without invalidating any theoretical properties of the original algorithm. <|reference_end|>", "<|reference_start|> Weight Uncertainty in Neural Networks: We introduce a new, efficient, principled and backpropagation-compatible algorithm for learning a probability distribution on the weights of a neural network, called Bayes by Backprop. It regularises the weights by minimising a compression cost, known as the variational free energy or the expected lower bound on the marginal likelihood. We show that this principled kind of regularisation yields comparable performance to dropout on MNIST classification. We then demonstrate how the learnt uncertainty in the weights can be used to improve generalisation in non-linear regression problems, and how this weight uncertainty can be used to drive the exploration-exploitation trade-off in reinforcement learning. <|reference_end|>", "<|reference_start|> Deep Exploration via Randomized Value Functions.: We study the use of randomized value functions to guide deep exploration in reinforcement learning. This offers an elegant means for synthesizing statistically and computationally efficient exploration with common practical approaches to value function learning. We present several reinforcement learning algorithms that leverage randomized value functions and demonstrate their efficacy through computational studies. We also prove a regret bound that establishes statistical efficiency with a tabular representation. <|reference_end|>" ]

[ 0, 10, 28, 44 ]

[ "<|reference_start|> Fast Approximate Clearance Evaluation for Rovers with Articulated Suspension Systems: We present a light-weight body-terrain clearance evaluation algorithm for the automated path planning of NASA's Mars 2020 rover. Extraterrestrial path planning is challenging due to the combination of terrain roughness and severe limitation in computational resources. Path planning on cluttered and/or uneven terrains requires repeated safety checks on all the candidate paths at a small interval. Predicting the future rover state requires simulating the vehicle settling on the terrain, which involves an inverse-kinematics problem with iterative nonlinear optimization under geometric constraints. However, such expensive computation is intractable for slow spacecraft computers, such as RAD750, which is used by the Curiosity Mars rover and upcoming Mars 2020 rover. We propose the Approximate Clearance Evaluation (ACE) algorithm, which obtains conservative bounds on vehicle clearance, attitude, and suspension angles without iterative computation. It obtains those bounds by estimating the lowest and highest heights that each wheel may reach given the underlying terrain, and calculating the worst-case vehicle configuration associated with those extreme wheel heights. The bounds are guaranteed to be conservative, hence ensuring vehicle safety during autonomous navigation. ACE is planned to be used as part of the new onboard path planner of the Mars 2020 rover. This paper describes the algorithm in detail and validates our claim of conservatism and fast computation through experiments. <|reference_end|>", "<|reference_start|> Using Occupancy Grids for Mobile Robot Perception and navigation: An approach to robot perception and world modeling that uses a probabilistic tesselated representation of spatial information called the occupancy grid is reviewed. The occupancy grid is a multidimensional random field that maintains stochastic estimates of the occupancy state of the cells in a spatial lattice. To construct a sensor-derived map of the robot's world, the cell state estimates are obtained by interpreting the incoming range readings using probabilistic sensor models. Bayesian estimation procedures allow the incremental updating of the occupancy grid, using readings taken from several sensors over multiple points of view. The use of occupancy grids from mapping and for navigation is examined. Operations on occupancy grids and extensions of the occupancy grid framework are briefly considered.<<ETX>> <|reference_end|>", "<|reference_start|> Design and comparative evaluation of an iterative contact point estimation method for static stability estimation of mobile actively reconfigurable robots: <|reference_end|>", "<|reference_start|> Pose Estimation of Vehicles Over Uneven Terrain: This paper presents a method for pose estimation of off-road vehicles moving over uneven terrain. It determines the contact points between the wheels and the terrain, assuming rigid contacts between an arbitrary number of wheels and ground. The terrain is represented by a 3D points cloud, interpolated by a B-patch to provide a continuous terrain representation. The pose estimation problem is formulated as a rigid body contact problem for a given location of the vehicle's center of mass over the terrain and a given yaw angle. The contact points between the wheels and ground are determined by releasing the vehicle from a given point above the terrain, until the contact forces between the wheels and ground, and the gravitational force, reach equilibrium. The contact forces are calculated using singular value decomposition (SVD) of the deduced contact matrix. The proposed method is computationally efficient, allowing real time computation during motion, as demonstrated in several examples. Accurate pose estimations can be used for motion planning, stability analyses and traversability analyses over uneven terrain. <|reference_end|>" ]

[ 1, 9, 11, 15 ]

[ "<|reference_start|> Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference: Deep learning has recently demonstrated its excellent performance for multi-view stereo (MVS). However, one major limitation of current learned MVS approaches is the scalability: the memory-consuming cost volume regularization makes the learned MVS hard to be applied to high-resolution scenes. In this paper, we introduce a scalable multi-view stereo framework based on the recurrent neural network. Instead of regularizing the entire 3D cost volume in one go, the proposed Recurrent Multi-view Stereo Network (R-MVSNet) sequentially regularizes the 2D cost maps along the depth direction via the gated recurrent unit (GRU). This reduces dramatically the memory consumption and makes high-resolution reconstruction feasible. We first show the state-of-the-art performance achieved by the proposed R-MVSNet on the recent MVS benchmarks. Then, we further demonstrate the scalability of the proposed method on several large-scale scenarios, where previous learned approaches often fail due to the memory constraint. Code is available at https://github.com/YoYo000/MVSNet. <|reference_end|>", "<|reference_start|> Multi-camera Scene Reconstruction via Graph Cuts: <|reference_end|>", "<|reference_start|> Unsupervised Learning of Depth and Ego-Motion from Video: We present an unsupervised learning framework for the task of monocular depth and camera motion estimation from unstructured video sequences. We achieve this by simultaneously training depth and camera pose estimation networks using the task of view synthesis as the supervisory signal. The networks are thus coupled via the view synthesis objective during training, but can be applied independently at test time. Empirical evaluation on the KITTI dataset demonstrates the effectiveness of our approach: 1) monocular depth performing comparably with supervised methods that use either ground-truth pose or depth for training, and 2) pose estimation performing favorably with established SLAM systems under comparable input settings. <|reference_end|>", "<|reference_start|> Pixelwise View Selection for Unstructured Multi-View Stereo: <|reference_end|>" ]

[ 13, 17, 56, 63 ]

[ "<|reference_start|> Mapping from frame-driven to frame-free event-driven vision systems by low-rate rate coding and coincidence processing--application to feedforward ConvNets.: Event-driven visual sensors have attracted interest from a number of different research communities. They provide visual information in quite a different way from conventional video systems consisting of sequences of still images rendered at a given \"frame rate.\" Event-driven vision sensors take inspiration from biology. Each pixel sends out an event (spike) when it senses something meaningful is happening, without any notion of a frame. A special type of event-driven sensor is the so-called dynamic vision sensor (DVS) where each pixel computes relative changes of light or \"temporal contrast.\" The sensor output consists of a continuous flow of pixel events that represent the moving objects in the scene. Pixel events become available with microsecond delays with respect to \"reality.\" These events can be processed \"as they flow\" by a cascade of event (convolution) processors. As a result, input and output event flows are practically coincident in time, and objects can be recognized as soon as the sensor provides enough meaningful events. In this paper, we present a methodology for mapping from a properly trained neural network in a conventional frame-driven representation to an event-driven representation. The method is illustrated by studying event-driven convolutional neural networks (ConvNet) trained to recognize rotating human silhouettes or high speed poker card symbols. The event-driven ConvNet is fed with recordings obtained from a real DVS camera. The event-driven ConvNet is simulated with a dedicated event-driven simulator and consists of a number of event-driven processing modules, the characteristics of which are obtained from individually manufactured hardware modules. <|reference_end|>", "<|reference_start|> Real-Time Classification and Sensor Fusion with a Spiking Deep Belief Network: Deep Belief Networks (DBNs) have recently shown impressive performance on a broad range of classification problems. Their generative properties allow better understanding of the performance, and provide a simpler solution for sensor fusion tasks. However, because of their inherent need for feedback and parallel update of large numbers of units, DBNs are expensive to implement on serial computers. This paper proposes a method based on the Siegert approximation for Integrate-and-Fire neurons to map an offline-trained DBN onto an efficient event-driven spiking neural network suitable for hardware implementation. The method is demonstrated in simulation and by a real-time implementation of a 3-layer network with 2694 neurons used for visual classification of MNIST handwritten digits with input from a 128 × 128 Dynamic Vision Sensor (DVS) silicon retina, and sensory-fusion using additional input from a 64-channel AER-EAR silicon cochlea. The system is implemented through the open-source software in the jAER project and runs in real-time on a laptop computer. It is demonstrated that the system can recognize digits in the presence of distractions, noise, scaling, translation and rotation, and that the degradation of recognition performance by using an event-based approach is less than 1%. Recognition is achieved in an average of 5.8 ms after the onset of the presentation of a digit. By cue integration from both silicon retina and cochlea outputs we show that the system can be biased to select the correct digit from otherwise ambiguous input. <|reference_end|>", "<|reference_start|> A 128$\\,\\times$ 128 1.5% Contrast Sensitivity 0.9% FPN 3 µs Latency 4 mW Asynchronous Frame-Free Dynamic Vision Sensor Using Transimpedance Preamplifiers: Dynamic Vision Sensors (DVS) have recently appeared as a new paradigm for vision sensing and processing. They feature unique characteristics such as contrast coding under wide illumination variation, micro-second latency response to fast stimuli, and low output data rates (which greatly improves the efficiency of post-processing stages). They can track extremely fast objects (e.g., time resolution is better than 100 kFrames/s video) without special lighting conditions. Their availability has triggered a new range of vision applications in the fields of surveillance, motion analyses, robotics, and microscopic dynamic observations. One key DVS feature is contrast sensitivity, which has so far been reported to be in the 10-15% range. In this paper, a novel pixel photo sensing and transimpedance pre-amplification stage makes it possible to improve by one order of magnitude contrast sensitivity (down to 1.5%) and power (down to 4 mW), reduce the best reported FPN (Fixed Pattern Noise) by a factor of 2 (down to 0.9%), while maintaining the shortest reported latency (3 μs) and good Dynamic Range (120 dB), and further reducing overall area (down to 30 × 31 μm per pixel). The only penalty is the limitation of intrascene Dynamic Range to 3 decades. A 128 × 128 DVS test prototype has been fabricated in standard 0.35 μm CMOS and extensive experimental characterization results are provided. <|reference_end|>", "<|reference_start|> Microsaccades: A microcosm for research on oculomotor control, attention, and visual perception.: <|reference_end|>" ]

[ 2, 10, 11, 12 ]

[ "<|reference_start|> Weak convergence and optimal scaling of random walk Metropolis algorithms: This paper considers the problem of scaling the proposal distribution of a multidimensional random walk Metropolis algorithm in order to maximize the efficiency of the algorithm. The main result is a weak convergence result as the dimension of a sequence of target densities, n, converges to `. When the proposal variance is appropriately scaled according to n, the sequence of stochastic processes formed by the first component of each Markov chain converges to the appropriate limiting Langevin diffusion process. The limiting diffusion approximation admits a straightforward efficiency maximization problem, and the resulting asymptotically optimal policy is related to the asymptotic acceptance rate of proposed moves for the algorithm. The asymptotically optimal acceptance rate is 0.234 under quite general conditions. The main result is proved in the case where the target density has a symmetric product form. Extensions of the result are discussed. <|reference_end|>", "<|reference_start|> Efficient sampling using metropolis algorithms: Applications of optimal scaling results: We recently considered the optimal scaling problem of Metropolis algorithms for multidimensional target distributions with non-IID components. The results that were proven have wide applications and the aim of this article is to show how practitioners can take advantage of them. In particular, we use several examples to illustrate the casewhere the asymptotically optimal acceptance rate is the usual 0.234, and also the latest developments where smaller acceptance rates should be adopted for optimal sampling from the target distributions involved. We study the impact of the proposal scaling on the performance of the algorithm, and finally perform simulation studies exploring the efficiency of the algorithm when sampling from some popular statistical models. <|reference_end|>", "<|reference_start|> International Conference on Learning Representations (ICLR): <|reference_end|>", "<|reference_start|> Auxiliary Variational MCMC: <|reference_end|>" ]

[ 1, 8, 20, 34 ]

[ "<|reference_start|> Image Transformer: Image generation has been successfully cast as an autoregressive sequence generation or transformation problem. Recent work has shown that self-attention is an effective way of modeling textual sequences. In this work, we generalize a recently proposed model architecture based on self-attention, the Transformer, to a sequence modeling formulation of image generation with a tractable likelihood. By restricting the self-attention mechanism to attend to local neighborhoods we significantly increase the size of images the model can process in practice, despite maintaining significantly larger receptive fields per layer than typical convolutional neural networks. While conceptually simple, our generative models significantly outperform the current state of the art in image generation on ImageNet, improving the best published negative log-likelihood on ImageNet from 3.83 to 3.77. We also present results on image super-resolution with a large magnification ratio, applying an encoder-decoder configuration of our architecture. In a human evaluation study, we find that images generated by our super-resolution model fool human observers three times more often than the previous state of the art. <|reference_end|>", "<|reference_start|> Rectifying the Shortcut Learning of Background for Few-Shot Learning: The category gap between training and evaluation has been characterised as one of the main obstacles to the success of Few-Shot Learning (FSL). In this paper, we for the first time empirically identify image background, common in realistic images, as a shortcut knowledge helpful for in-class classification but ungeneralizable beyond training categories in FSL. A novel framework, COSOC, is designed to tackle this problem by extracting foreground objects in images at both training and evaluation without any extra supervision. Extensive experiments carried on inductive FSL tasks demonstrate the effectiveness of our approaches. <|reference_end|>", "<|reference_start|> Three things everyone should know about Vision Transformers: After their initial success in natural language processing, transformer architectures have rapidly gained traction in computer vision, providing state-of-the-art results for tasks such as image classification, detection, segmentation, and video analysis. We offer three insights based on simple and easy to implement variants of vision transformers. (1) The residual layers of vision transformers, which are usually processed sequentially, can to some extent be processed efficiently in parallel without noticeably affecting the accuracy. (2) Fine-tuning the weights of the attention layers is sufficient to adapt vision transformers to a higher resolution and to other classification tasks. This saves compute, reduces the peak memory consumption at fine-tuning time, and allows sharing the majority of weights across tasks. (3) Adding MLP-based patch pre-processing layers improves Bert-like self-supervised training based on patch masking. We evaluate the impact of these design choices using the ImageNet-1k dataset, and confirm our findings on the ImageNet-v2 test set. Transfer performance is measured across six smaller datasets. <|reference_end|>", "<|reference_start|> Rethinking ImageNet Pre-training: We report competitive results on object detection and instance segmentation on the COCO dataset using standard models trained from random initialization. The results are no worse than their ImageNet pre-training counterparts even when using the hyper-parameters of the baseline system (Mask R-CNN) that were optimized for fine-tuning pre-trained models, with the sole exception of increasing the number of training iterations so the randomly initialized models may converge. Training from random initialization is surprisingly robust; our results hold even when: (i) using only 10% of the training data, (ii) for deeper and wider models, and (iii) for multiple tasks and metrics. Experiments show that ImageNet pre-training speeds up convergence early in training, but does not necessarily provide regularization or improve final target task accuracy. To push the envelope we demonstrate 50.9 AP on COCO object detection without using any external data---a result on par with the top COCO 2017 competition results that used ImageNet pre-training. These observations challenge the conventional wisdom of ImageNet pre-training for dependent tasks and we expect these discoveries will encourage people to rethink the current de facto paradigm of `pre-training and fine-tuning' in computer vision. <|reference_end|>" ]

[ 14, 28, 29, 34 ]

[ "<|reference_start|> Energy Efficient Legitimate Wireless Surveillance of UAV Communications: Unmanned aerial vehicles (UAVs) enhance connectivity and accessibility for civilian and military applications. Criminals or terrorists can potentially use UAVs for committing crimes and terrorism, thus endangering public safety. In this paper, we consider that a legitimate UAV is employed to track flight of suspicious UAVs for preventing safety and security threats. To obtain flight information of the suspicious UAVs, the legitimate UAV intentionally jams the suspicious receiver so as to force the suspicious UAV to reduce its data rate, and hence increase the eavesdropping success. An energy-efficient jamming strategy is proposed for the legitimate UAV to maximize the amount of eavesdropped packets. Moreover, a tracking algorithm is developed for the legitimate UAV to track the suspicious flight by comprehensively utilizing eavesdropped packets, angle-of-arrival and received signal strength of the suspicious transmitter's signal. A new simulation framework is implemented to combine the complementary features of optimization toolbox with channel modeling (in MATLAB) and discrete event-driven mobility tracking (in NS3). Moreover, numerical results validate the proposed algorithms in terms of packet eavesdropping rate and tracking accuracy of the suspicious UAVs’ trajectory. <|reference_end|>", "<|reference_start|> Poster: May the Swarm Be With You: Sensor Spoofing Attacks Against Drone Swarms: Swarm robotics, particularly drone swarms, are used in various safety-critical tasks. While a lot of attention has been paid to improving swarm control algorithms for improved intelligence, the security implications of various design choices in swarm control algorithms have not been studied. We highlight how an attacker can exploit the vulnerabilities in swarm control algorithms to disrupt drone swarms. Specifically, we show that the attacker can target one swarm member (target drone) through sensor spoofing attacks, and indirectly cause other swarm members (victim drones) to veer off from their course, and potentially resulting in a crash. Our attack cannot be prevented by traditional software security techniques, and it is stealthy in nature as it causes seemingly benign deviations in drone swarms. Our initial results show that spoofing the position of a target drone by 5m is sufficient to cause other drones to crash into a front obstacle. Overall, our attack achieves 76.67% and 93.33% success rate with 5m and 10m spoofing deviation respectively. <|reference_end|>", "<|reference_start|> Detecting Attacks Against Robotic Vehicles: A Control Invariant Approach: Robotic vehicles (RVs), such as drones and ground rovers, are a type of cyber-physical systems that operate in the physical world under the control of computing components in the cyber world. Despite RVs' robustness against natural disturbances, cyber or physical attacks against RVs may lead to physical malfunction and subsequently disruption or failure of the vehicles' missions. To avoid or mitigate such consequences, it is essential to develop attack detection techniques for RVs. In this paper, we present a novel attack detection framework to identify external, physical attacks against RVs on the fly by deriving and monitoring Control Invariants (CI). More specifically, we propose a method to extract such invariants by jointly modeling a vehicle's physical properties, its control algorithm and the laws of physics. These invariants are represented in a state-space form, which can then be implemented and inserted into the vehicle's control program binary for runtime invariant check. We apply our CI framework to eleven RVs, including quadrotor, hexarotor, and ground rover, and show that the invariant check can detect three common types of physical attacks -- including sensor attack, actuation signal attack, and parameter attack -- with very low runtime overhead. <|reference_end|>", "<|reference_start|> Robust Localization for Secure Navigation of UAV Formations Under GNSS Spoofing Attack: Nowadays, aerial formations are frequently employed in outdoor scenarios to cooperatively explore and monitor wide areas of interest. In these applications, the vehicles are often exposed to relevant security vulnerabilities, as, for instance, the alteration of navigation signals from an attacker with map counterfeiting (if not even hijacking) purposes. In this work, we focus on an Unmanned Aerial Vehicle (UAV) formation that monitors an area, wherein navigation spoofing attacks may occur. Letting the UAVs cooperate and exploiting the redundancy in the available sensing information, a distributed procedure is proposed to $i$ ) detect spoofing attacks, and $ii$ ) support the navigation in adverse conditions. The validity of the designed approach is confirmed by numerical results. Aerial vehicles for outdoor operation are generally endowed with inertial measurements, relative ranging, and GNSS sensing capability. In this work, two cascaded estimation algorithms for concurrent GNSS spoofing detection and localization in a multi-UAV scenario is proposed, to attain robust navigation in areas subject to GNSS spoofing attacks. The attack detection leverages on information theoretic tools to provide a practical threshold test by checking the multimodal measurement consistency. The localization procedures exploit a decision logic relying on measurement reliability to combine information sources that are different in nature, for UAV self-localization in both safe and under-attack conditions. Note to Practitioners—Aerial vehicles for outdoor operation are generally endowed with inertial measurements, relative ranging, and GNSS sensing capability. In this work, two cascaded estimation algorithms for concurrent GNSS spoofing detection and localization in a multi-UAV scenario is proposed, to attain robust navigation in areas subject to GNSS spoofing attacks. The attack detection leverages on information theoretic tools to provide a practical threshold test by checking the multimodal measurement consistency. The localization procedures exploit a decision logic relying on measurement reliability to combine information sources that are different in nature, for UAV self-localization in both safe and under-attack conditions. <|reference_end|>" ]

[ 4, 10, 11, 20 ]

[ "<|reference_start|> DRONE: Data-aware Low-rank Compression for Large NLP Models: The representations learned by large-scale NLP models such as BERT have been widely used in various tasks. However, the increasing model size of the pre-trained models also brings efficiency challenges, including inference speed and model size when deploying models on mobile devices. Specifically, most operations in BERT consist of matrix multiplications. These matrices are not low-rank and thus canonical matrix decompositions do not lead to efficient approximations. In this paper, we observe that the learned representation of each layer lies in a low-dimensional space. Based on this observation, we propose DRONE ( d ata-awa r e l o w-ra n k compr e ssion), a provably optimal low-rank decomposition of weight matrices, which has a simple closed form solution that can be efficiently computed. DRONE can be applied to both fully-connected and self-attention layers appearing in the BERT model. In addition to compressing standard models, our method can also be used on distilled BERT models to further improve the compression rate. Experimental results show that DRONE is able to improve both model size and inference speed with limited loss in accuracy. Specifically, DRONE alone achieves 1.92x speedup on the MRPC task with only 1.5 % loss in accuracy, and when DRONE is combined with distillation, it further achieves over 12.3x speedup on various natural language inference tasks. <|reference_end|>", "<|reference_start|> GroupReduce: Block-Wise Low-Rank Approximation for Neural Language Model Shrinking: Model compression is essential for serving large deep neural nets on devices with limited resources or applications that require real-time responses. As a case study, a state-of-the-art neural language model usually consists of one or more recurrent layers sandwiched between an embedding layer used for representing input tokens and a softmax layer for generating output tokens. For problems with a very large vocabulary size, the embedding and the softmax matrices can account for more than half of the model size. For instance, the bigLSTM model achieves state-of- the-art performance on the One-Billion-Word (OBW) dataset with around 800k vocabulary, and its word embedding and softmax matrices use more than 6GBytes space, and are responsible for over 90% of the model parameters. In this paper, we propose GroupReduce, a novel compression method for neural language models, based on vocabulary-partition (block) based low-rank matrix approximation and the inherent frequency distribution of tokens (the power-law distribution of words). The experimental results show our method can significantly outperform traditional compression methods such as low-rank approximation and pruning. On the OBW dataset, our method achieved 6.6 times compression rate for the embedding and softmax matrices, and when combined with quantization, our method can achieve 26 times compression rate, which translates to a factor of 12.8 times compression for the entire model with very little degradation in perplexity. <|reference_end|>", "<|reference_start|> Attention is Not All You Need: Pure Attention Loses Rank Doubly Exponentially with Depth: Attention-based architectures have become ubiquitous in machine learning, yet our understanding of the reasons for their effectiveness remains limited. This work proposes a new way to understand self-attention networks: we show that their output can be decomposed into a sum of smaller terms, each involving the operation of a sequence of attention heads across layers. Using this decomposition, we prove that self-attention possesses a strong inductive bias towards \"token uniformity\". Specifically, without skip connections or multi-layer perceptrons (MLPs), the output converges doubly exponentially to a rank-1 matrix. On the other hand, skip connections and MLPs stop the output from degeneration. Our experiments verify the identified convergence phenomena on different variants of standard transformer architectures. <|reference_end|>", "<|reference_start|> Understanding Batch Normalization: Batch normalization (BN) is a technique to normalize activations in intermediate layers of deep neural networks. Its tendency to improve accuracy and speed up training have established BN as a favorite technique in deep learning. Yet, despite its enormous success, there remains little consensus on the exact reason and mechanism behind these improvements. In this paper we take a step towards a better understanding of BN, following an empirical approach. We conduct several experiments, and show that BN primarily enables training with larger learning rates, which is the cause for faster convergence and better generalization. For networks without BN we demonstrate how large gradient updates can result in diverging loss and activations growing uncontrollably with network depth, which limits possible learning rates. BN avoids this problem by constantly correcting activations to be zero-mean and of unit standard deviation, which enables larger gradient steps, yields faster convergence and may help bypass sharp local minima. We further show various ways in which gradients and activations of deep unnormalized networks are ill-behaved. We contrast our results against recent findings in random matrix theory, shedding new light on classical initialization schemes and their consequences. <|reference_end|>" ]

[ 10, 23, 27, 29 ]

[ "<|reference_start|> Simple Data Augmentation for Multilingual NLU in Task Oriented Dialogue Systems: Data augmentation has shown potential in alleviating data scarcity for Natural Language Understanding (e.g. slot filling and intent classification) in task-oriented dialogue systems. As prior work has been mostly experimented on English datasets, we focus on five different languages, and consider a setting where limited data are available. We investigate the effectiveness of non-gradient based augmentation methods, involving simple text span substitutions and syntactic manipulations. Our experiments show that (i) augmentation is effective in all cases, particularly for slot filling; and (ii) it is beneficial for a joint intent-slot model based on multilingual BERT, both for limited data settings and when full training data is used. <|reference_end|>", "<|reference_start|> Improving Robustness of Task Oriented Dialog Systems: Task oriented language understanding in dialog systems is often modeled using intents (task of a query) and slots (parameters for that task). Intent detection and slot tagging are, in turn, modeled using sentence classification and word tagging techniques respectively. Similar to adversarial attack problems with computer vision models discussed in existing literature, these intent-slot tagging models are often over-sensitive to small variations in input -- predicting different and often incorrect labels when small changes are made to a query, thus reducing their accuracy and reliability. However, evaluating a model's robustness to these changes is harder for language since words are discrete and an automated change (e.g. adding `noise') to a query sometimes changes the meaning and thus labels of a query. In this paper, we first describe how to create an adversarial test set to measure the robustness of these models. Furthermore, we introduce and adapt adversarial training methods as well as data augmentation using back-translation to mitigate these issues. Our experiments show that both techniques improve the robustness of the system substantially and can be combined to yield the best results. <|reference_end|>", "<|reference_start|> GOLD: Improving Out-of-Scope Detection in Dialogues using Data Augmentation: Practical dialogue systems require robust methods of detecting out-of-scope (OOS) utterances to avoid conversational breakdowns and related failure modes. Directly training a model with labeled OOS examples yields reasonable performance, but obtaining such data is a resource-intensive process. To tackle this limited-data problem, previous methods focus on better modeling the distribution of in-scope (INS) examples. We introduce GOLD as an orthogonal technique that augments existing data to train better OOS detectors operating in low-data regimes. GOLD generates pseudo-labeled candidates using samples from an auxiliary dataset and keeps only the most beneficial candidates for training through a novel filtering mechanism. In experiments across three target benchmarks, the top GOLD model outperforms all existing methods on all key metrics, achieving relative gains of 52.4%, 48.9% and 50.3% against median baseline performance. We also analyze the unique properties of OOS data to identify key factors for optimally applying our proposed method. <|reference_end|>", "<|reference_start|> Data Augmentation for Spoken Language Understanding via Joint Variational Generation: Data scarcity is one of the main obstacles of domain adaptation in spoken language understanding (SLU) due to the high cost of creating manually tagged SLU datasets. Recent works in neural text generative models, particularly latent variable models such as variational autoencoder (VAE), have shown promising results in regards to generating plausible and natural sentences. In this paper, we propose a novel generative architecture which leverages the generative power of latent variable models to jointly synthesize fully annotated utterances. Our experiments show that existing SLU models trained on the additional synthetic examples achieve performance gains. Our approach not only helps alleviate the data scarcity issue in the SLU task for many datasets but also indiscriminately improves language understanding performances for various SLU models, supported by extensive experiments and rigorous statistical testing. <|reference_end|>" ]

[ 3, 6, 7, 11 ]

[ "<|reference_start|> Decentralising finance using decentralised blockchain oracles: The recent spread of COVID-19, stressed economies and government pumping money into the market has once again ignited the discussion on the need to have decentralised economies, the role of regulatory authorities and if bitcoin represents a true store of value. In this paper, we identify the need to alternate financial structure, discuss how blockchain and cryptocurrencies play a very important role in achieving it. Blockchain applications are heavily dependent on oracles for their interaction with outside world, we have discussed here the functioning of oracles and then finally presented a broad architecture that can be used to implement a majority of financial instruments on blockchain. <|reference_end|>", "<|reference_start|> Visa: 随着互联网技术和通信技术的快速发展，消费者的消费习惯和支付方式也发生了巨大的变化。Visa中国区副总经理、创新产品负责入龚亮告诉记者，整个支付行业正在经历快速的变革,必须要有更先进的平台和服务才能跟上需求变化的步伐。 <|reference_end|>", "<|reference_start|> Ouroboros: A Provably Secure Proof-of-Stake Blockchain Protocol: <|reference_end|>", "<|reference_start|> Plonk: Permutations over lagrange-bases for oecumenical\nnoninteractive arguments of knowledge: zk-SNARK constructions that utilize an updatable universal structured reference string remove one of the main obstacles in deploying zk-SNARKs[GKM + ]. The important work of Maller et al. [MBKM19] presented Sonic - the first potentially practical zk-SNARK with fully succinct verification for general arithmetic circuits with such an SRS. However, the version of Sonic enabling fully succinct verification still requires relatively high proof construction overheads. We present a universal SNARK construction with fully succinct verification, and significantly lower prover running time (roughly 7.5-20 times fewer group exponentiations than [MBKM19] in the fully succinct verifier mode depending on circuit structure). Similarly to [MBKM19] we rely on a permutation argument based on Bayer and Groth [BG12]. However, we focus on “Evaluations on a subgroup rather than coefficients of monomials”; which enables simplifying both the permutation argument and the arithmetization step. <|reference_end|>" ]

[ 0, 7, 8, 17 ]

[ "<|reference_start|> A distributed dual ascent algorithm for the Hop-constrained Steiner Tree Problem: <|reference_end|>", "<|reference_start|> The Steiner tree problem with hop constraints: <|reference_end|>", "<|reference_start|> MIP models for connected facility location: A theoretical and computational study☆: <|reference_end|>", "<|reference_start|> A distributed dual ascent algorithm for the Hop-constrained Steiner Tree Problem: <|reference_end|>" ]

[ 4, 8, 10, 12 ]

[ "<|reference_start|> An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: While the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision, attention is either applied in conjunction with convolutional networks, or used to replace certain components of convolutional networks while keeping their overall structure in place. We show that this reliance on CNNs is not necessary and a pure transformer applied directly to sequences of image patches can perform very well on image classification tasks. When pre-trained on large amounts of data and transferred to multiple mid-sized or small image recognition benchmarks (ImageNet, CIFAR-100, VTAB, etc.), Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train. <|reference_end|>", "<|reference_start|> Advances and Open Problems in Federated Learning: Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges. <|reference_end|>", "<|reference_start|> Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding: Neural networks are both computationally intensive and memory intensive, making them difficult to deploy on embedded systems with limited hardware resources. To address this limitation, we introduce \"deep compression\", a three stage pipeline: pruning, trained quantization and Huffman coding, that work together to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy. Our method first prunes the network by learning only the important connections. Next, we quantize the weights to enforce weight sharing, finally, we apply Huffman coding. After the first two steps we retrain the network to fine tune the remaining connections and the quantized centroids. Pruning, reduces the number of connections by 9x to 13x; Quantization then reduces the number of bits that represent each connection from 32 to 5. On the ImageNet dataset, our method reduced the storage required by AlexNet by 35x, from 240MB to 6.9MB, without loss of accuracy. Our method reduced the size of VGG-16 by 49x from 552MB to 11.3MB, again with no loss of accuracy. This allows fitting the model into on-chip SRAM cache rather than off-chip DRAM memory. Our compression method also facilitates the use of complex neural networks in mobile applications where application size and download bandwidth are constrained. Benchmarked on CPU, GPU and mobile GPU, compressed network has 3x to 4x layerwise speedup and 3x to 7x better energy efficiency. <|reference_end|>", "<|reference_start|> Communication-Efficient Learning of Deep Networks from Decentralized Data: Modern mobile devices have access to a wealth of data suitable for learning models, which in turn can greatly improve the user experience on the device. For example, language models can improve speech recognition and text entry, and image models can automatically select good photos. However, this rich data is often privacy sensitive, large in quantity, or both, which may preclude logging to the data center and training there using conventional approaches. We advocate an alternative that leaves the training data distributed on the mobile devices, and learns a shared model by aggregating locally-computed updates. We term this decentralized approach Federated Learning. We present a practical method for the federated learning of deep networks based on iterative model averaging, and conduct an extensive empirical evaluation, considering five different model architectures and four datasets. These experiments demonstrate the approach is robust to the unbalanced and non-IID data distributions that are a defining characteristic of this setting. Communication costs are the principal constraint, and we show a reduction in required communication rounds by 10-100x as compared to synchronized stochastic gradient descent. <|reference_end|>" ]

[ 1, 3, 5, 14 ]

[ "<|reference_start|> The complexity of homomorphism and constraint satisfaction problems\nseen from the other side: We give a complexity theoretic classification of homomorphism problems for graphs and, more generally, relational structures obtained by restricting the left hand side structure in a homomorphism. For every class C of structures, let HOM(C, /spl I.bar/) be the problem of deciding whether a given structure A /spl isin/ C has a homomorphism to a given (arbitrary) structure B. We prove that, under some complexity theoretic assumption from parameterized complexity theory, HOM(C, /spl I.bar/) is in polynomial time if, and only if, the cores of all structures in C have bounded tree-width (as long as the structures in C only contain relations of bounded arity). Due to a well known correspondence between homomorphism problems and constraint satisfaction problems, our classification carries over to the latter. <|reference_end|>", "<|reference_start|> Learning Description Logic Concepts: When can Positive and Negative Examples be Separated? (Abstract): Learning description logic (DL) concepts from positive and negative examples given in the form of labeled data items in a KB has received significant attention in the literature. We study the fundamental question of when a separating DL concept exists and provide useful model-theoretic characterizations as well as complexity results for the associated decision problem. For expressive DLs such as ALC and ALCQI, our characterizations show a surprising link to the evaluation of ontology-mediated conjunctive queries. We exploit this to determine the combined complexity (between ExpTime and NExpTime) and data complexity (second level of the polynomial hierarchy) of separability. For the Horn DL EL, separability is ExpTime-complete both in combined and in data complexity while for its modest extension ELI it is even undecidable. Separability is also undecidable when the KB is formulated in ALC and the separating concept is required to be in EL or ELI. <|reference_end|>", "<|reference_start|> Learning Conjunctive Concepts in Structural Domains: <|reference_end|>", "<|reference_start|> A Dichotomy Theorem for Learning Quantified Boolean Formulas: <|reference_end|>" ]

[ 1, 10, 20, 29 ]

[ "<|reference_start|> Synthetic generation of radio maps for device-free passive localization: In this paper, we present the design, implementation, and evaluation of a system that automatically constructs accurate radio maps for device-free WLAN localization systems. The system is capable of generating deterministic and probabilistic radio maps for localization systems. Our system uses 3D ray tracing enhanced with the uniform theory of diffraction (UTD) to model the electric field behavior and the human shadowing effect. We present our system architecture and describe the details of its different components. We also propose an optional module, location-0 correction, that can significantly enhances the system accuracy and reduces its dependence on the 3D model details by using just one signal strength sample. Our experiments in a real testbed show that the predicted signal strength differs from the measurements by a maximum average absolute error of $2.77$ dB achieving a maximum localization error of $3.13$m and $2.84$m for both the deterministic and probabilistic radio maps, respectively. In addition, the results show that our system is not sensitive to the 3D model details. <|reference_end|>", "<|reference_start|> An improved localization algorithm in wireless sensor network: An improved localization algorithm W-Ecolocation in wireless sensor network is proposed. When there is a high erroneous constraints rate, the Ecolocation algorithm which only considers the locations with maximum computing number of matched constraints cannot obtain the optimal location estimate. The W-Ecolocation makes use of more locations with great computing number of matched constraints and takes the weighted average as the location estimate of the unknown node. The simulation results show that W-Ecolocation performs better than Ecolocation in terms of location error and location precision. A new approach to compute the number of matched constraints is also presented. It reduces both the time complexity and space complexity. <|reference_end|>", "<|reference_start|> Implications of device diversity for organic localization: Many indoor localization methods are based on the association of 802.11 wireless RF signals from wireless access points (WAPs) with location labels. An “organic” RF positioning system relies on regular users, not dedicated surveyors, to build the map of RF fingerprints to location labels. However, signal variation due to device heterogeneity may degrade localization performance. We analyze the diversity of those signal characteristics pertinent to indoor localization — signal strength and AP detection — as measured by a variety of 802.11 devices. We first analyze signal strength diversity, and show that pairwise linear transformation alone does not solve the problem. We propose kernel estimation with a wide kernel width to reduce the difference in probability estimates. We also investigate diversity in access point detection. We demonstrate that localization performance may degrade significantly when AP detection rate is used as a feature for localization, and correlate the loss of performance to a device dissimilarity measure captured by Kullback-Leibler divergence. Based on this analysis, we show that using only signal strength, without incorporating negative evidence, achieves good localization performance when devices are heterogeneous. <|reference_end|>", "<|reference_start|> Covariate Shift in Hilbert Space: A Solution via Sorrogate Kernels: Covariate shift is an unconventional learning scenario in which training and testing data have different distributions. A general principle to solve the problem is to make the training data distribution similar to that of the test domain, such that classifiers computed on the former generalize well to the latter. Current approaches typically target on sample distributions in the input space, however, for kernel-based learning methods, the algorithm performance depends directly on the geometry of the kernel-induced feature space. Motivated by this, we propose to match data distributions in the Hilbert space, which, given a pre-defined empirical kernel map, can be formulated as aligning kernel matrices across domains. In particular, to evaluate similarity of kernel matrices defined on arbitrarily different samples, the novel concept of surrogate kernel is introduced based on the Mercer's theorem. Our approach caters the model adaptation specifically to kernel-based learning mechanism, and demonstrates promising results on several real-world applications. <|reference_end|>" ]

[ 1, 6, 16, 18 ]

[ "<|reference_start|> StarCoder: may the source be with you!: The BigCode community, an open-scientific collaboration working on the responsible development of Large Language Models for Code (Code LLMs), introduces StarCoder and StarCoderBase: 15.5B parameter models with 8K context length, infilling capabilities and fast large-batch inference enabled by multi-query attention. StarCoderBase is trained on 1 trillion tokens sourced from The Stack, a large collection of permissively licensed GitHub repositories with inspection tools and an opt-out process. We fine-tuned StarCoderBase on 35B Python tokens, resulting in the creation of StarCoder. We perform the most comprehensive evaluation of Code LLMs to date and show that StarCoderBase outperforms every open Code LLM that supports multiple programming languages and matches or outperforms the OpenAI code-cushman-001 model. Furthermore, StarCoder outperforms every model that is fine-tuned on Python, can be prompted to achieve 40\\% pass@1 on HumanEval, and still retains its performance on other programming languages. We take several important steps towards a safe open-access model release, including an improved PII redaction pipeline and a novel attribution tracing tool, and make the StarCoder models publicly available under a more commercially viable version of the Open Responsible AI Model license. <|reference_end|>", "<|reference_start|> Evaluating Large Language Models Trained on Code: We introduce Codex, a GPT language model fine-tuned on publicly available code from GitHub, and study its Python code-writing capabilities. A distinct production version of Codex powers GitHub Copilot. On HumanEval, a new evaluation set we release to measure functional correctness for synthesizing programs from docstrings, our model solves 28.8% of the problems, while GPT-3 solves 0% and GPT-J solves 11.4%. Furthermore, we find that repeated sampling from the model is a surprisingly effective strategy for producing working solutions to difficult prompts. Using this method, we solve 70.2% of our problems with 100 samples per problem. Careful investigation of our model reveals its limitations, including difficulty with docstrings describing long chains of operations and with binding operations to variables. Finally, we discuss the potential broader impacts of deploying powerful code generation technologies, covering safety, security, and economics. <|reference_end|>", "<|reference_start|> IntelliCode Compose: Code Generation Using Transformer: In software development through integrated development environments (IDEs), code completion is one of the most widely used features. Nevertheless, majority of integrated development environments only support completion of methods and APIs, or arguments. In this paper, we introduce IntelliCode Compose $-$ a general-purpose multilingual code completion tool which is capable of predicting sequences of code tokens of arbitrary types, generating up to entire lines of syntactically correct code. It leverages state-of-the-art generative transformer model trained on 1.2 billion lines of source code in Python, $C\\#$, JavaScript and TypeScript programming languages. IntelliCode Compose is deployed as a cloud-based web service. It makes use of client-side tree-based caching, efficient parallel implementation of the beam search decoder, and compute graph optimizations to meet edit-time completion suggestion requirements in the Visual Studio Code IDE and Azure Notebook. Our best model yields an average edit similarity of $86.7\\%$ and a perplexity of 1.82 for Python programming language. <|reference_end|>", "<|reference_start|> Training language models to follow instructions with human feedback: Making language models bigger does not inherently make them better at following a user's intent. For example, large language models can generate outputs that are untruthful, toxic, or simply not helpful to the user. In other words, these models are not aligned with their users. In this paper, we show an avenue for aligning language models with user intent on a wide range of tasks by fine-tuning with human feedback. Starting with a set of labeler-written prompts and prompts submitted through the OpenAI API, we collect a dataset of labeler demonstrations of the desired model behavior, which we use to fine-tune GPT-3 using supervised learning. We then collect a dataset of rankings of model outputs, which we use to further fine-tune this supervised model using reinforcement learning from human feedback. We call the resulting models InstructGPT. In human evaluations on our prompt distribution, outputs from the 1.3B parameter InstructGPT model are preferred to outputs from the 175B GPT-3, despite having 100x fewer parameters. Moreover, InstructGPT models show improvements in truthfulness and reductions in toxic output generation while having minimal performance regressions on public NLP datasets. Even though InstructGPT still makes simple mistakes, our results show that fine-tuning with human feedback is a promising direction for aligning language models with human intent. <|reference_end|>" ]

[ 7, 8, 11, 19 ]

[ "<|reference_start|> Generating Long Sequences with Sparse Transformers: Transformers are powerful sequence models, but require time and memory that grows quadratically with the sequence length. In this paper we introduce sparse factorizations of the attention matrix which reduce this to $O(n \\sqrt{n})$. We also introduce a) a variation on architecture and initialization to train deeper networks, b) the recomputation of attention matrices to save memory, and c) fast attention kernels for training. We call networks with these changes Sparse Transformers, and show they can model sequences tens of thousands of timesteps long using hundreds of layers. We use the same architecture to model images, audio, and text from raw bytes, setting a new state of the art for density modeling of Enwik8, CIFAR-10, and ImageNet-64. We generate unconditional samples that demonstrate global coherence and great diversity, and show it is possible in principle to use self-attention to model sequences of length one million or more. <|reference_end|>", "<|reference_start|> Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention: Transformers achieve remarkable performance in several tasks but due to their quadratic complexity, with respect to the input's length, they are prohibitively slow for very long sequences. To address this limitation, we express the self-attention as a linear dot-product of kernel feature maps and make use of the associativity property of matrix products to reduce the complexity from $\\mathcal{O}\\left(N^2\\right)$ to $\\mathcal{O}\\left(N\\right)$, where $N$ is the sequence length. We show that this formulation permits an iterative implementation that dramatically accelerates autoregressive transformers and reveals their relationship to recurrent neural networks. Our linear transformers achieve similar performance to vanilla transformers and they are up to 4000x faster on autoregressive prediction of very long sequences. <|reference_end|>", "<|reference_start|> Rethinking Attention with Performers: We introduce Performers, Transformer architectures which can estimate regular (softmax) full-rank-attention Transformers with provable accuracy, but using only linear (as opposed to quadratic) space and time complexity, without relying on any priors such as sparsity or low-rankness. To approximate softmax attention-kernels, Performers use a novel Fast Attention Via positive Orthogonal Random features approach (FAVOR+), which may be of independent interest for scalable kernel methods. FAVOR+ can be also used to efficiently model kernelizable attention mechanisms beyond softmax. This representational power is crucial to accurately compare softmax with other kernels for the first time on large-scale tasks, beyond the reach of regular Transformers, and investigate optimal attention-kernels. Performers are linear architectures fully compatible with regular Transformers and with strong theoretical guarantees: unbiased or nearly-unbiased estimation of the attention matrix, uniform convergence and low estimation variance. We tested Performers on a rich set of tasks stretching from pixel-prediction through text models to protein sequence modeling. We demonstrate competitive results with other examined efficient sparse and dense attention methods, showcasing effectiveness of the novel attention-learning paradigm leveraged by Performers. <|reference_end|>", "<|reference_start|> Linformer: Self-Attention with Linear Complexity: Large transformer models have shown extraordinary success in achieving state-of-the-art results in many natural language processing applications. However, training and deploying these models can be prohibitively costly for long sequences, as the standard self-attention mechanism of the Transformer uses $O(n^2)$ time and space with respect to sequence length. In this paper, we demonstrate that the self-attention mechanism can be approximated by a low-rank matrix. We further exploit this finding to propose a new self-attention mechanism, which reduces the overall self-attention complexity from $O(n^2)$ to $O(n)$ in both time and space. The resulting linear transformer, the \\textit{Linformer}, performs on par with standard Transformer models, while being much more memory- and time-efficient. <|reference_end|>" ]

[ 3, 10, 20, 43 ]

[ "<|reference_start|> Quaternionic periodicity transform: an algebraic solution to the tandem repeat detection problem: MOTIVATION\nOne of the main tasks of DNA sequence analysis is identification of repetitive patterns. DNA symbol repetitions play a key role in a number of applications, including prediction of gene and exon locations, identification of diseases, reconstruction of human evolutionary history and DNA forensics.\n\n\nRESULTS\nA new approach towards identification of tandem repeats in DNA sequences is proposed. The approach is a refinement of previously considered method, based on the complex periodicity transform. The refinement is obtained, among others, by mapping of DNA symbols to pure quaternions. This mapping results in an enhanced, symbol-balanced sensitivity of the transform to DNA patterns, and an unambiguous threshold selection criterion. Computational efficiency of the transform is further improved, and coupling of the computation with the period value is removed, thereby facilitating parallel implementation of the algorithm. Additionally, a post-processing stage is inserted into the algorithm, enabling unambiguous display of results in a convenient graphical format. Comparison of the quaternionic periodicity transform with two well-known pattern detection techniques shows that the new approach is competitive with these two techniques in detection of exact and approximate repeats. <|reference_end|>", "<|reference_start|> Introduction to the Modern Theory of Dynamical Systems: <|reference_end|>", "<|reference_start|> Digital pulse processing in high resolution, high throughput, gamma-ray spectroscopy: A new method for processing signals produced by high resolution, large volume semiconductor detectors is described. These detectors, to be used in the next generation of spectrometer arrays for nuclear research (i.e., EUROBALL, etc.), present a set of problems, such as resolution degradation due to charge trapping and ballistic deficit effects, poor resolution at a high count rate, long term and temperature instability, etc. To solve these problems, a new approach based on digital moving window deconvolution (MWD) is developed. > <|reference_end|>", "<|reference_start|> The complexity of the minimum k-cover problem: The k-coversproblem (kCP asks us to compute a minimum cardinality set of strings given length k>1 that covers a given string. It was shown in a recent paper, by reduction to 3 -SAT, that the k-covers problem is NP-complete. In this paper we introduce a new problem, that we call the Relaxed Vertex Cover Problem (RVCP), which we show is a special case of Set Cover (SCP). We show further the kCP is equivalent to RVCP restricted to certain classes GXk of graphs that represent all strings x. We discuss approximate solutions of kCP and we state a number of conjectures and open problems related to kCP and GXk. <|reference_end|>" ]

[ 6, 7, 10, 25 ]

[ "<|reference_start|> {DC Optimal Power Flow with Joint Chance Constraints: Managing uncertainty and variability in power injections has become a major concern for power system operators due to increasing levels of fluctuating renewable energy connected to the grid. This work addresses this uncertainty via a joint chance-constrained formulation of the DC optimal power flow (OPF) problem, which satisfies all the constraints jointly with a pre-determined probability. The few existing approaches for solving joint chance-constrained OPF problems are typically either computationally intractable for large-scale problems or give overly conservative solutions that satisfy the constraints far more often than required, resulting in excessively costly operation. This paper proposes an algorithm for solving joint chance-constrained DC OPF problems by adopting an S$\\ell _1$QP-type trust-region algorithm. This algorithm uses a sample-based approach that avoids making strong assumptions on the distribution of the uncertainties, scales favorably to large problems, and can be tuned to obtain less conservative results. We illustrate the performance of our method using several IEEE test cases. The results demonstrate the proposed algorithm's advantages in computational times and limited conservativeness of the solutions relative to other joint chance-constrained DC OPF algorithms. <|reference_end|>", "<|reference_start|> Chance-Constrained Model Predictive Control for Drinking Water Networks: <|reference_end|>", "<|reference_start|> Efficient Relaxations for Joint Chance Constrained AC Optimal Power Flow: <|reference_end|>", "<|reference_start|> Efficient Relaxations for Joint Chance Constrained AC Optimal Power Flow: <|reference_end|>" ]

[ 2, 5, 6, 8 ]

[ "<|reference_start|> World-Scale Mining of Objects and Events from Community Photo Collections: In this paper, we describe an approach for mining images of objects (such as touristic sights) from community photo collections in an unsupervised fashion. Our approach relies on retrieving geotagged photos from those web-sites using a grid of geospatial tiles. The downloaded photos are clustered into potentially interesting entities through a processing pipeline of several modalities, including visual, textual and spatial proximity. The resulting clusters are analyzed and are automatically classified into objects and events. Using mining techniques, we then find text labels for these clusters, which are used to again assign each cluster to a corresponding Wikipedia article in a fully unsupervised manner. A final verification step uses the contents (including images) from the selected Wikipedia article to verify the cluster-article assignment. We demonstrate this approach on several urban areas, densely covering an area of over 700 square kilometers and mining over 200,000 photos, making it probably the largest experiment of its kind to date. <|reference_end|>", "<|reference_start|> Large-Scale Location Recognition And The Geometric Burstiness Problem: Visual location recognition is the task of determining the place depicted in a query image from a given database of geo-tagged images. Location recognition is often cast as an image retrieval problem and recent research has almost exclusively focused on improving the chance that a relevant database image is ranked high enough after retrieval. The implicit assumption is that the number of inliers found by spatial verification can be used to distinguish between a related and an unrelated database photo with high precision. In this paper, we show that this assumption does not hold for large datasets due to the appearance of geometric bursts, i.e., sets of visual elements appearing in similar geometric configurations in unrelated database photos. We propose algorithms for detecting and handling geometric bursts. Although conceptually simple, using the proposed weighting schemes dramatically improves the recall that can be achieved when high precision is required compared to the standard re-ranking based on the inlier count. Our approach is easy to implement and can easily be integrated into existing location recognition systems. <|reference_end|>", "<|reference_start|> Improving Image-Based Localization by Active Correspondence Search: <|reference_end|>", "<|reference_start|> Geolocating static cameras: A key problem in widely distributed camera networks is locating the cameras. This paper considers three scenarios for camera localization: localizing a camera in an unknown environment, adding a new camera in a region with many other cameras, and localizing a camera by finding correlations with satellite imagery. We find that simple summary statistics (the time course of principal component coefficients) are sufficient to geolocate cameras without determining correspondences between cameras or explicitly reasoning about weather in the scene. We present results from a database of images from 538 cameras collected over the course of a year. We find that for cameras that remain stationary and for which we have accurate image times- tamps, we can localize most cameras to within 50 miles of the known location. In addition, we demonstrate the use of a distributed camera network in the construction a map of weather conditions. <|reference_end|>" ]

[ 21, 25, 34, 42 ]

[ "<|reference_start|> Probabilistic Skyline Operator over Sliding Windows: <|reference_end|>", "<|reference_start|> Who solved the secretary problem.: In Martin Gardner's Mathematical Games column in the February 1960 issue of Scientific American, there appeared a simple problem that has come to be known today as the Secretary Problem, or the Marriage Problem. It has since been taken up and developed by many eminent probabilists and statisticians and has been extended and generalized in many different directions so that now one can say that it constitutes a \"field\" within mathematics-probability-optimization. The object of this article is partly historical (to give a fresh view of the origins of the problem, touching upon Cayley and Kepler), partly review of the field (listing the subfields of recent interest), partly serious (to answer the question posed in the title), and partly entertainment. The contents of this paper were first given as the Allen T. Craig lecture at the University of Iowa, 1988. <|reference_end|>", "<|reference_start|> Universal Approximations for TSP, Steiner Tree, and Set Cover: We introduce a notion of universality in the context of optimization problems with partial information. Universality is a framework for dealing with uncertainty by guaranteeing a certain quality of goodness for all possible completions of the partial information set. Universal variants of optimization problems can be defined that are both natural and well-motivated. We consider universal versions of three classical problems: TSP, Steiner Tree and Set Cover.We present a polynomial-time algorithm to find a universal tour on a given metric space over n vertices such that for any subset of the vertices, the sub-tour induced by the subset is within O(log4n/log log n) of an optimal tour for the subset. Similarly, we show that given a metric space over n vertices and a root vertex, we can find a universal spanning tree such that for any subset of vertices containing the root, the sub-tree induced by the subset is within O(log4n/log log n) of an optimal Steiner tree for the subset. Our algorithms rely on a new notion of sparse partitions, that may be of independent interest. For the special case of doubling metrics, which includes both constant-dimensional Euclidean and growth-restricted metrics, our algorithms achieve an O(log n) upper bound. We complement our results for the universal Steiner tree problem with a lower bound of Ω(log n/log log n) that holds even for n vertices on the plane. We also show that a slight generalization of the universal Steiner Tree problem is coNP-hard and present nearly tight upper and lower bounds for a universal version of Set Cover. <|reference_end|>", "<|reference_start|> Optimal Oblivious Routing in Polynomial Time: A recent seminal result of Racke is that for any network there is an oblivious routing algorithm with a polylog competitive ratio with respect to congestion. Unfortunately, Racke's construction is not polynomial time. We give a polynomial time construction that guarantee's Racke's bounds, and more generally gives the true optimal ratio for any network. <|reference_end|>" ]

[ 7, 8, 12, 13 ]

[ "<|reference_start|> Continual Occlusion and Optical Flow Estimation: <|reference_end|>", "<|reference_start|> UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Loss: In the era of end-to-end deep learning, many advances in computer vision are driven by large amounts of labeled data. In the optical flow setting, however, obtaining dense per-pixel ground truth for real scenes is difficult and thus such data is rare. Therefore, recent end-to-end convolutional networks for optical flow rely on synthetic datasets for supervision, but the domain mismatch between training and test scenarios continues to be a challenge. Inspired by classical energy-based optical flow methods, we design an unsupervised loss based on occlusion-aware bidirectional flow estimation and the robust census transform to circumvent the need for ground truth flow. On the KITTI benchmarks, our unsupervised approach outperforms previous unsupervised deep networks by a large margin, and is even more accurate than similar supervised methods trained on synthetic datasets alone. By optionally fine-tuning on the KITTI training data, our method achieves competitive optical flow accuracy on the KITTI 2012 and 2015 benchmarks, thus in addition enabling generic pre-training of supervised networks for datasets with limited amounts of ground truth. <|reference_end|>", "<|reference_start|> PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume: We present a compact but effective CNN model for optical flow, called PWC-Net. PWC-Net has been designed according to simple and well-established principles: pyramidal processing, warping, and the use of a cost volume. Cast in a learnable feature pyramid, PWC-Net uses the cur- rent optical flow estimate to warp the CNN features of the second image. It then uses the warped features and features of the first image to construct a cost volume, which is processed by a CNN to estimate the optical flow. PWC-Net is 17 times smaller in size and easier to train than the recent FlowNet2 model. Moreover, it outperforms all published optical flow methods on the MPI Sintel final pass and KITTI 2015 benchmarks, running at about 35 fps on Sintel resolution (1024x436) images. Our models are available on https://github.com/NVlabs/PWC-Net. <|reference_end|>", "<|reference_start|> A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation: Recent work has shown that optical flow estimation can be formulated as a supervised learning task and can be successfully solved with convolutional networks. Training of the so-called FlowNet was enabled by a large synthetically generated dataset. The present paper extends the concept of optical flow estimation via convolutional networks to disparity and scene flow estimation. To this end, we propose three synthetic stereo video datasets with sufficient realism, variation, and size to successfully train large networks. Our datasets are the first large-scale datasets to enable training and evaluating scene flow methods. Besides the datasets, we present a convolutional network for real-time disparity estimation that provides state-of-the-art results. By combining a flow and disparity estimation network and training it jointly, we demonstrate the first scene flow estimation with a convolutional network. <|reference_end|>" ]

[ 7, 20, 28, 37 ]

[ "<|reference_start|> Traffic Congestion Detection from Surveillance Videos using Deep Learning: Countless cameras, both public and private, have been installed in recent years for the objectives of surveillance, the monitoring of anomalous human activities, and traffic surveillance. Numerous worrisome and aberrant actions, such as theft, aggression, and accidents, make it difficult to notice and recognise such behaviour in a real-world setting. The topic of this study is car wrecks as depicted in online videos of traffic. Modern traffic monitoring and surveillance rely heavily on video traffic surveillance cameras (VTSS). Consequences of a rapidly expanding human population include a higher frequency of accidental injuries. The VTSS is employed to identify unusual occurrences on various roads and highways, such as traffic congestion and car accidents. When accidents happen on lengthy roadways or in remote areas, victims are often powerless and some don't make it. The purpose of this study is to provide a method for automatically identifying incidents in surveillance footage. Convolutional-neural-networks (CNNs), a specific deep learning approach developed to cope with grid-like data, have been shown to be useful in image and video processing, according to a study of the relevant literature. This study use a rolling prediction method and convolutional neural networks (CNNs) to detect accidents in VTSS footage. A dataset of anomalous photographs, called the Vehicle Accident Image Dataset (VAID), was created and used in the training of the CNN model. The proposed method was put through its paces by analysing data gathered from running the trained CNN model on a number of different films. This study's findings demonstrate a 93% success rate in identifying traffic accident incidents in films from traffic surveillance systems. <|reference_end|>", "<|reference_start|> Video Anomaly Detection for Pedestrian Surveillance: <|reference_end|>", "<|reference_start|> Traffic object detection and recognition based on the attentional visual field of drivers: Traffic object detection and recognition systems play an essential role in Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles (AV). In this research, we focus on four important classes of traffic objects: traffic signs, road vehicles, pedestrians, and traffic lights. We first review the major traditional machine learning and deep learning methods that have been used in the literature to detect and recognize these objects. We provide a vision-based framework that detects and recognizes traffic objects inside and outside the attentional visual area of drivers. This approach uses the driver 3D absolute coordinates of the gaze point obtained by the combined, cross-calibrated use of a front-view stereo imaging system and a non-contact 3D gaze tracker. A combination of multi-scale HOG-SVM and Faster R-CNN-based models are utilized in the detection stage. The recognition stage is performed with a ResNet-101 network to verify sets of generated hypotheses. We applied our approach on real data collected during drives in an urban environment with the RoadLAB instrumented vehicle. Our framework achieved 91% of correct object detections and provided promising results in the object recognition stage. <|reference_end|>", "<|reference_start|> An Extendable, Efficient and Effective Transformer-based Object Detector: Transformers have been widely used in numerous vision problems especially for visual recognition and detection. Detection transformers are the first fully end-to-end learning systems for object detection, while vision transformers are the first fully transformer-based architecture for image classification. In this paper, we integrate Vision and Detection Transformers (ViDT) to construct an effective and efficient object detector. ViDT introduces a reconfigured attention module to extend the recent Swin Transformer to be a standalone object detector, followed by a computationally efficient transformer decoder that exploits multi-scale features and auxiliary techniques essential to boost the detection performance without much increase in computational load. In addition, we extend it to ViDT+ to support joint-task learning for object detection and instance segmentation. Specifically, we attach an efficient multi-scale feature fusion layer and utilize two more auxiliary training losses, IoU-aware loss and token labeling loss. Extensive evaluation results on the Microsoft COCO benchmark dataset demonstrate that ViDT obtains the best AP and latency trade-off among existing fully transformer-based object detectors, and its extended ViDT+ achieves 53.2AP owing to its high scalability for large models. The source code and trained models are available at https://github.com/naver-ai/vidt. <|reference_end|>" ]

[ 10, 11, 22, 36 ]

[ "<|reference_start|> Wide & Deep Learning for Recommender Systems: Generalized linear models with nonlinear feature transformations are widely used for large-scale regression and classification problems with sparse inputs. Memorization of feature interactions through a wide set of cross-product feature transformations are effective and interpretable, while generalization requires more feature engineering effort. With less feature engineering, deep neural networks can generalize better to unseen feature combinations through low-dimensional dense embeddings learned for the sparse features. However, deep neural networks with embeddings can over-generalize and recommend less relevant items when the user-item interactions are sparse and high-rank. In this paper, we present Wide & Deep learning---jointly trained wide linear models and deep neural networks---to combine the benefits of memorization and generalization for recommender systems. We productionized and evaluated the system on Google Play, a commercial mobile app store with over one billion active users and over one million apps. Online experiment results show that Wide & Deep significantly increased app acquisitions compared with wide-only and deep-only models. We have also open-sourced our implementation in TensorFlow. <|reference_end|>", "<|reference_start|> On Using Very Large Target Vocabulary for Neural Machine Translation: Neural machine translation, a recently proposed approach to machine translation based purely on neural networks, has shown promising results compared to the existing approaches such as phrase-based statistical machine translation. Despite its recent success, neural machine translation has its limitation in handling a larger vocabulary, as training complexity as well as decoding complexity increase proportionally to the number of target words. In this paper, we propose a method that allows us to use a very large target vocabulary without increasing training complexity, based on importance sampling. We show that decoding can be efficiently done even with the model having a very large target vocabulary by selecting only a small subset of the whole target vocabulary. The models trained by the proposed approach are empirically found to outperform the baseline models with a small vocabulary as well as the LSTM-based neural machine translation models. Furthermore, when we use the ensemble of a few models with very large target vocabularies, we achieve the state-of-the-art translation performance (measured by BLEU) on the English->German translation and almost as high performance as state-of-the-art English->French translation system. <|reference_end|>", "<|reference_start|> Locating targets through mention in Twitter: <|reference_end|>", "<|reference_start|> Mention recommendation in Twitter with cooperative multi-agent reinforcement learning: In Twitter-like social networking services, the \"@'' symbol can be used with the tweet to mention users whom the user wants to alert regarding the message. An automatic suggestion to the user of a small list of candidate names can improve communication efficiency. Previous work usually used several most recent tweets or randomly select historical tweets to make an inference about this preferred list of names. However, because there are too many historical tweets by users and a wide variety of content types, the use of several tweets cannot guarantee the desired results. In this work, we propose the use of a novel cooperative multi-agent approach to mention recommendation, which incorporates dozens of more historical tweets than earlier approaches. The proposed method can effectively select a small set of historical tweets and cooperatively extract relevant indicator tweets from both the user and mentioned users. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods. <|reference_end|>" ]

[ 11, 16, 24, 25 ]

[ "<|reference_start|> Buckets of resistance: Standards and the effectiveness of citizen science: In light of arguments that citizen science has the potential to make environmental knowledge and policy more robust and democratic, this article inquires into the factors that shape the ability of citizen science to actually influence scientists and decision makers. Using the case of community-based air toxics monitoring with ‘‘buckets,’’ it argues that citizen science’s effectiveness is significantly influenced by standards and standardized practices. It demonstrates that, on one hand, standards serve a boundary-bridging function that affords bucket monitoring data a crucial measure of legitimacy among experts. On the other hand, standards simultaneously serve a boundary-policing function, allowing experts to dismiss bucket data as irrelevant to the central project of air quality assessment. The article thus calls attention to standard setting as an important site of intervention for citizen science-based efforts to democratize science and policy. <|reference_end|>", "<|reference_start|> Citizen Scientists: Reconnecting Science with Civil Society: <|reference_end|>", "<|reference_start|> Designing Interactive Systems for Community Citizen Science: Citizen science forges partnerships between experts and citizens through collaboration and has become a trend in public participation in scientific research over the past decade. Besides this trend, public participation can also contribute to participatory democracy, which empowers citizens to advocate for their local problems. This strategy supports citizens to form a community, increase environmental monitoring, gather evidence, and tell convincing stories. Researchers believe that this “community citizen science” strategy can contribute to the well-being of communities by giving them the power to influence the general public and decision makers. Community citizen science requires collecting, curating, visualizing, analyzing, and interpreting multiple types of data over a large spacetime scale. This is highly dependent on community engagement (i.e., the involvement of citizens in local neighborhoods). Such large-scale tasks require the assistance of innovative computational tools to give technology affordance to communities. However, existing tools often focus on only one type of data, and thus researchers need to develop tools from scratch. Moreover, there is a lack of design patterns for researchers to reference when developing such tools. Furthermore, existing tools are typically treated as products rather than ongoing infrastructures that sustain community engagement. This research studies the methodology of developing computational tools by using visualization, crowdsourcing, and artificial intelligence techniques to support the entire community engagement lifecycle, from initiation, maintenance, to evaluation. This research will make methodological and empirical contributions to community citizen science and sustainable human-computer interaction. Methodological contributions include detailed case studies with applied techniques from information technology systems that are deployed in real-world contexts. Empirical contributions include generalizable empirical insights for developing interactive systems that integrate multiple types of scientific data. In this dissertation, I first define “community citizen science” and explain corresponding design challenges. Then, I review existing computational tools and techniques that are related to this research. Next, I present four interactive systems centered around the research scope: (1) a timelapse editor that supports building evidence-based narratives, (2) an air quality monitoring system that integrates heterogeneous data and computer vision to support the formation of scientific knowledge, (3) a visualization tool that reveals the impact of oil and gas development, and (4) a mobile crowdsourced application for reporting and visualizing pollution odors. Finally, I synthesize findings from all four works into generalizable design implications for future researchers and developers. <|reference_end|>", "<|reference_start|> A Low-tech Sensing System for Particulate Pollution: We present an ultra low-cost sensing system, which enables participants to see and reflect on the particulates in their air. Drawing on prior work in paper computing, we introduce small sensors for particulate pollution that can be easily assembled from common paper materials for less than $1 USD, and mailed by regular postal service to residents of entire neighborhoods, cities, or geographic regions. Recipients collect particulate samples using these sensors and mail them back to a central location, where the particles are viewed and analyzed via a microscope. The data, which includes rich images of actual air pollution particles, can then be broadcast to larger audiences. This paper details the design of our system and its deployment with a local air quality activist community. We conclude by highlighting the tradeoffs between high-tech and low-tech sensing, and suggest opportunities for tangible interaction to support rich, new ways of seeing our environment. <|reference_end|>" ]

[ 9, 26, 32, 53 ]

[ "<|reference_start|> Segment Anything: We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate its capabilities on numerous tasks and find that its zero-shot performance is impressive -- often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at https://segment-anything.com to foster research into foundation models for computer vision. <|reference_end|>", "<|reference_start|> Segment Anything in High Quality: The recent Segment Anything Model (SAM) represents a big leap in scaling up segmentation models, allowing for powerful zero-shot capabilities and flexible prompting. Despite being trained with 1.1 billion masks, SAM's mask prediction quality falls short in many cases, particularly when dealing with objects that have intricate structures. We propose HQ-SAM, equipping SAM with the ability to accurately segment any object, while maintaining SAM's original promptable design, efficiency, and zero-shot generalizability. Our careful design reuses and preserves the pre-trained model weights of SAM, while only introducing minimal additional parameters and computation. We design a learnable High-Quality Output Token, which is injected into SAM's mask decoder and is responsible for predicting the high-quality mask. Instead of only applying it on mask-decoder features, we first fuse them with early and final ViT features for improved mask details. To train our introduced learnable parameters, we compose a dataset of 44K fine-grained masks from several sources. HQ-SAM is only trained on the introduced detaset of 44k masks, which takes only 4 hours on 8 GPUs. We show the efficacy of HQ-SAM in a suite of 10 diverse segmentation datasets across different downstream tasks, where 8 out of them are evaluated in a zero-shot transfer protocol. Our code and pretrained models are at https://github.com/SysCV/SAM-HQ. <|reference_end|>", "<|reference_start|> Customized Segment Anything Model for Medical Image Segmentation: We propose SAMed, a general solution for medical image segmentation. Different from the previous methods, SAMed is built upon the large-scale image segmentation model, Segment Anything Model (SAM), to explore the new research paradigm of customizing large-scale models for medical image segmentation. SAMed applies the low-rank-based (LoRA) finetuning strategy to the SAM image encoder and finetunes it together with the prompt encoder and the mask decoder on labeled medical image segmentation datasets. We also observe the warmup finetuning strategy and the AdamW optimizer lead SAMed to successful convergence and lower loss. Different from SAM, SAMed could perform semantic segmentation on medical images. Our trained SAMed model achieves 81.88 DSC and 20.64 HD on the Synapse multi-organ segmentation dataset, which is on par with the state-of-the-art methods. We conduct extensive experiments to validate the effectiveness of our design. Since SAMed only updates a small fraction of the SAM parameters, its deployment cost and storage cost are quite marginal in practical usage. The code of SAMed is available at https://github.com/hitachinsk/SAMed. <|reference_end|>", "<|reference_start|> Exemplar Learning for Medical Image Segmentation: Medical image annotation typically requires expert knowledge and hence incurs time-consuming and expensive data annotation costs. To alleviate this burden, we propose a novel learning scenario, Exemplar Learning (EL), to explore automated learning processes for medical image segmentation with a single annotated image example. This innovative learning task is particularly suitable for medical image segmentation, where all categories of organs can be presented in one single image and annotated all at once. To address this challenging EL task, we propose an Exemplar Learning-based Synthesis Net (ELSNet) framework for medical image segmentation that enables innovative exemplar-based data synthesis, pixel-prototype based contrastive embedding learning, and pseudo-label based exploitation of the unlabeled data. Specifically, ELSNet introduces two new modules for image segmentation: an exemplar-guided synthesis module, which enriches and diversifies the training set by synthesizing annotated samples from the given exemplar, and a pixel-prototype based contrastive embedding module, which enhances the discriminative capacity of the base segmentation model via contrastive representation learning. Moreover, we deploy a two-stage process for segmentation model training, which exploits the unlabeled data with predicted pseudo segmentation labels. To evaluate this new learning framework, we conduct extensive experiments on several organ segmentation datasets and present an in-depth analysis. The empirical results show that the proposed exemplar learning framework produces effective segmentation results. <|reference_end|>" ]

[ 3, 8, 12, 19 ]

[ "<|reference_start|> Codebook Design for Millimeter-Wave Channel Estimation with Hybrid Precoding Structure: In this paper, we study hierarchical codebook design for channel estimation in millimeter-wave (mmWave) communications with a hybrid precoding structure. Due to the limited saturation power of mmWave power amplifier (PA), we take the per-antenna power constraint (PAPC) into consideration. We first propose a metric, i.e., generalized detection probability (GDP), to evaluate the quality of \\emph{an arbitrary codeword}. This metric not only enables an optimization approach for mmWave codebook design, but also can be used to compare the performance of two different codewords/codebooks. To the best of our knowledge, GDP is the first metric particularly for mmWave codebook design for channel estimation. We then propose an approach to design a hierarchical codebook exploiting BeaM Widening with Multi-RF-chain Sub-array technique (BMW-MS). To obtain crucial parameters of BMW-MS, we provide two solutions, namely a low-complexity search (LCS) solution to optimize the GDP metric and a closed-form (CF) solution to pursue a flat beam pattern. Performance comparisons show that BMW-MS/LCS and BMW-MS/CF achieve very close performances, and they outperform the existing alternatives under the PAPC. <|reference_end|>", "<|reference_start|> Compressive channel estimation and tracking for large arrays in mm-wave picocells: We propose and investigate a compressive architecture for estimation and tracking of sparse spatial channels in millimeter (mm) wave picocellular networks. The base stations are equipped with antenna arrays with a large number of elements (which can fit within compact form factors because of the small carrier wavelength) and employ radio frequency (RF) beamforming, so that standard least squares adaptation techniques (which require access to individual antenna elements) are not applicable. We focus on the downlink, and show that “compressive beacons,” transmitted using pseudorandom phase settings at the base station array, and compressively processed using pseudorandom phase settings at the mobile array, provide information sufficient for accurate estimation of the two-dimensional (2D) spatial frequencies associated with the directions of departure of the dominant rays from the base station, and the associated complex gains. This compressive approach is compatible with coarse phase-only control, and is based on a near-optimal sequential algorithm for frequency estimation which approaches the Cramér Rao Lower Bound. The algorithm exploits the geometric continuity of the channel across successive beaconing intervals to reduce the overhead to less than 1% even for very large (32 × 32) arrays. Compressive beaconing is essentially omnidirectional, and hence does not enjoy the SNR and spatial reuse benefits of beamforming obtained during data transmission. We therefore discuss system level design considerations for ensuring that the beacon SNR is sufficient for accurate channel estimation, and that inter-cell beacon interference is controlled by an appropriate reuse scheme. <|reference_end|>", "<|reference_start|> {Delay-Phase Precoding for THz Massive MIMO with Beam Split: Benefiting from tens of GHz bandwidth, Terahertz (THz) communications has been considered as one of the promising technologies for the future 6G wireless communications. To compensate the serious attenuation in THz band and avoid huge power consumption, massive multiple input multiple output (MIMO) with hybrid precoding is widely considered. However, the traditional phase-shifter (PS) based hybrid precoding architecture cannot cope with the effect of beam split in THz communications, which means that the path components of THz channel split into different spatial directions at different subcarrier frequencies, leading serious array gain loss. In this paper, we first point out the seriousness of beam split effect in THz massive MIMO by analyzing the array gain loss caused by the beam split effect. To compensate this array gain loss, we propose a new hybrid precoding architecture called delay-phase precoding (DPP). In the proposed DPP, a time delay (TD) network is introduced between radio- frequency chains and the traditional PS network, which converts phase-controlled analog precoding into delay-phase controlled analog precoding. When carrying out precoding, the time delays in the TD network are dedicatedly designed to generate frequency-dependent beams which are aligned with the spatial directions over the whole bandwidth. Thanks to the joint control of delay and phase, the proposed DPP can significantly alleviate the beam split effect. Simulation results reveal that the proposed DPP can generate beams with the near- optimal array gain over the whole bandwidth, and achieve the near-optimal achievable rate performance. <|reference_end|>", "<|reference_start|> Estimation of wideband dynamic mmWave and THz channels for 5G systems and beyond: Millimeter wave (mmWave) wideband channels in a multiple-input multiple-output (MIMO) transmission are described by a sparse set of impulse responses in the angle-delay, or space-time (ST), domain. These characteristics will be even more prominent in the THz band used in future systems. We consider two approaches for channel estimation: compressed-sensing (CS), exploiting the sparsity in the angular/delay domain, and low-rank (LR), exploiting the algebraic structure of channel matrix. Both approaches share several commonalities, and this paper provides for the first time i) a comparison of the two approaches, and ii) new versions of CS and LR methods that significantly improve performance in terms of mean squared error (MSE), computational complexity, and latency. We derive the asymptotic MSE bound for any estimator of the ST-MIMO multipath channels with invariant angles/delays and time-varying fading, with unknown angle/delay diversity order: the bound also accounts for the degradation introduced by sub-optimal separable channel models. We will show that in the considered scenarios both CS and LR approaches attain the bound. Our performance assessment over ideal and $3^{rd}$ generation partnership project (3GPP) channel models, suitable for the fifth-generation (5G) and beyond of cellular networks, shows the trade-off obtained by the methods over various metrics: i) CS methods are converging faster than the LR methods, both attaining the asymptotic MSE bound; ii) the CS methods depend on the array manifold, while LR methods are independent of the array calibration; iii) CS solutions are more complex than LR solutions. <|reference_end|>" ]

[ 11, 31, 38, 42 ]

[ "<|reference_start|> Recipes for building an open-domain chatbot: Building open-domain chatbots is a challenging area for machine learning research. While prior work has shown that scaling neural models in the number of parameters and the size of the data they are trained on gives improved results, we show that other ingredients are important for a high-performing chatbot. Good conversation requires a number of skills that an expert conversationalist blends in a seamless way: providing engaging talking points and listening to their partners, and displaying knowledge, empathy and personality appropriately, while maintaining a consistent persona. We show that large scale models can learn these skills when given appropriate training data and choice of generation strategy. We build variants of these recipes with 90M, 2.7B and 9.4B parameter models, and make our models and code publicly available. Human evaluations show our best models are superior to existing approaches in multi-turn dialogue in terms of engagingness and humanness measurements. We then discuss the limitations of this work by analyzing failure cases of our models. <|reference_end|>", "<|reference_start|> DIRECTOR: Generator-Classifiers For Supervised Language Modeling: Current language models achieve low perplexity but their resulting generations still suffer from toxic responses, repetitiveness and contradictions. The standard language modeling setup fails to address these issues. In this paper, we introduce a new architecture, {\\sc Director}, that consists of a unified generator-classifier with both a language modeling and a classification head for each output token. Training is conducted jointly using both standard language modeling data, and data labeled with desirable and undesirable sequences. Experiments in several settings show that the model has competitive training and decoding speed compared to standard language models while yielding superior results, alleviating known issues while maintaining generation quality. It also outperforms existing model guiding approaches in terms of both accuracy and efficiency. <|reference_end|>", "<|reference_start|> Training language models to follow instructions with human feedback: Making language models bigger does not inherently make them better at following a user's intent. For example, large language models can generate outputs that are untruthful, toxic, or simply not helpful to the user. In other words, these models are not aligned with their users. In this paper, we show an avenue for aligning language models with user intent on a wide range of tasks by fine-tuning with human feedback. Starting with a set of labeler-written prompts and prompts submitted through the OpenAI API, we collect a dataset of labeler demonstrations of the desired model behavior, which we use to fine-tune GPT-3 using supervised learning. We then collect a dataset of rankings of model outputs, which we use to further fine-tune this supervised model using reinforcement learning from human feedback. We call the resulting models InstructGPT. In human evaluations on our prompt distribution, outputs from the 1.3B parameter InstructGPT model are preferred to outputs from the 175B GPT-3, despite having 100x fewer parameters. Moreover, InstructGPT models show improvements in truthfulness and reductions in toxic output generation while having minimal performance regressions on public NLP datasets. Even though InstructGPT still makes simple mistakes, our results show that fine-tuning with human feedback is a promising direction for aligning language models with human intent. <|reference_end|>", "<|reference_start|> {Toward an Architecture for Never-Ending Language Learning: We consider here the problem of building a never-ending language learner; that is, an intelligent computer agent that runs forever and that each day must (1) extract, or read, information from the web to populate a growing structured knowledge base, and (2) learn to perform this task better than on the previous day. In particular, we propose an approach and a set of design principles for such an agent, describe a partial implementation of such a system that has already learned to extract a knowledge base containing over 242,000 beliefs with an estimated precision of 74% after running for 67 days, and discuss lessons learned from this preliminary attempt to build a never-ending learning agent. <|reference_end|>" ]

[ 4, 9, 14, 21 ]

[ "<|reference_start|> {Differential privacy and robust statistics: We show by means of several examples that robust statistical estimators present an excellent starting point for differentially private estimators. Our algorithms use a new paradigm for differentially private mechanisms, which we call Propose-Test-Release (PTR), and for which we give a formal definition and general composition theorems. <|reference_end|>", "<|reference_start|> Concentrated Differential Privacy: Simplifications, Extensions, and Lower Bounds: \"Concentrated differential privacy\" was recently introduced by Dwork and Rothblum as a relaxation of differential privacy, which permits sharper analyses of many privacy-preserving computations. We present an alternative formulation of the concept of concentrated differential privacy in terms of the Renyi divergence between the distributions obtained by running an algorithm on neighboring inputs. With this reformulation in hand, we prove sharper quantitative results, establish lower bounds, and raise a few new questions. We also unify this approach with approximate differential privacy by giving an appropriate definition of \"approximate concentrated differential privacy.\" <|reference_end|>", "<|reference_start|> Privacy integrated queries: an extensible platform for privacy-preserving\ndata analysis: We report on the design and implementation of the Privacy Integrated Queries (PINQ) platform for privacy-preserving data analysis. PINQ provides analysts with a programming interface to unscrubbed data through a SQL-like language. At the same time, the design of PINQ's analysis language and its careful implementation provide formal guarantees of differential privacy for any and all uses of the platform. PINQ's unconditional structural guarantees require no trust placed in the expertise or diligence of the analysts, substantially broadening the scope for design and deployment of privacy-preserving data analysis, especially by non-experts. <|reference_end|>", "<|reference_start|> Lua---an extensible extension language: This paper describes Lua, a language for extending applications. Lua combines procedural features with powerful data description facilities, by using a simple, yet powerful, mechanism of tables . This mechanism implements the concepts of records, arrays and recursive data types (pointers), and adds some object‐oriented facilities, such as methods with dynamic dispatching. <|reference_end|>" ]

[ 12, 16, 17, 18 ]

[ "<|reference_start|> A Simpler Approach to Matrix Completion: This paper provides the best bounds to date on the number of randomly sampled entries required to reconstruct an unknown low rank matrix. These results improve on prior work by Candes and Recht, Candes and Tao, and Keshavan, Montanari, and Oh. The reconstruction is accomplished by minimizing the nuclear norm, or sum of the singular values, of the hidden matrix subject to agreement with the provided entries. If the underlying matrix satisfies a certain incoherence condition, then the number of entries required is equal to a quadratic logarithmic factor times the number of parameters in the singular value decomposition. The proof of this assertion is short, self contained, and uses very elementary analysis. The novel techniques herein are based on recent work in quantum information theory. <|reference_end|>", "<|reference_start|> Compressed sensing, sparse approximation, and low-rank matrix estimation: The importance of sparse signal structures has been recognized in a plethora of applications ranging from medical imaging to group disease testing to radar technology. It has been shown in practice that various signals of interest may be (approximately) sparsely modeled, and that sparse modeling is often beneficial, or even indispensable to signal recovery. Alongside an increase in applications, a rich theory of sparse and compressible signal recovery has recently been developed under the names compressed sensing (CS) and sparse approximation (SA). This revolutionary research has demonstrated that many signals can be recovered from severely undersampled measurements by taking advantage of their inherent low-dimensional structure. More recently, an offshoot of CS and SA has been a focus of research on other low-dimensional signal structures such as matrices of low rank. Low-rank matrix recovery (LRMR) is demonstrating a rapidly growing array of important applications such as quantum state tomography, triangulation from incomplete distance measurements, recommender systems (e.g., the Netflix problem), and system identification and control. In this dissertation, we examine CS, SA, and LRMR from a theoretical perspective. We consider a variety of different measurement and signal models, both random and deterministic, and mainly ask two questions. How many measurements are necessary? How large is the recovery error? We give theoretical lower bounds for both of these questions, including oracle and minimax lower bounds for the error. However, the main emphasis of the thesis is to demonstrate the efficacy of convex optimization---in particular l1 and nuclear-norm minimization based programs---in CS, SA, and LRMR. We derive upper bounds for the number of measurements required and the error derived by convex optimization, which in many cases match the lower bounds up to constant or logarithmic factors. The majority of these results do not require the restricted isometry property (RIP), a ubiquitous condition in the literature. <|reference_end|>", "<|reference_start|> An accelerated gradient method for trace norm minimization: We consider the minimization of a smooth loss function regularized by the trace norm of the matrix variable. Such formulation finds applications in many machine learning tasks including multi-task learning, matrix classification, and matrix completion. The standard semidefinite programming formulation for this problem is computationally expensive. In addition, due to the non-smooth nature of the trace norm, the optimal first-order black-box method for solving such class of problems converges as O(1/√k), where k is the iteration counter. In this paper, we exploit the special structure of the trace norm, based on which we propose an extended gradient algorithm that converges as O(1/k). We further propose an accelerated gradient algorithm, which achieves the optimal convergence rate of O(1/k2) for smooth problems. Experiments on multi-task learning problems demonstrate the efficiency of the proposed algorithms. <|reference_end|>", "<|reference_start|> Structured regularizers for high-dimensional problems: Statistical and computational issues: Regularization is a widely used technique throughout statistics, machine learning, and applied mathematics. Modern applications in science and engineering lead to massive and complex data sets, which motivate the use of more structured types of regularizers. This survey provides an overview of the use of structured regularization in high-dimensional statistics, including regularizers for group-structured and hierarchical sparsity, low-rank matrices, additive and multiplicative matrix decomposition, and high-dimensional nonparametric models. It includes various examples with motivating applications; it also covers key aspects of statistical theory and provides some discussion of efficient algorithms. 233 A nn ua l R ev ie w o f St at is tic s an d It s A pp lic at io n 20 14 .1 :2 33 -2 53 . D ow nl oa de d fr om w w w .a nn ua lr ev ie w s. or g by $ {i nd iv id ua lU se r. di sp la yN am e} o n 01 /0 9/ 14 . F or p er so na l u se o nl y. ST01CH11-Wainwright ARI 29 November 2013 14:44 <|reference_end|>" ]

[ 9, 28, 32, 33 ]

[ "<|reference_start|> Learning representations by back-propagating errors: <|reference_end|>", "<|reference_start|> Modeling Musical Context with Word2vec: We present a semantic vector space model for capturing complex polyphonic musical context. A word2vec model based on a skip-gram representation with negative sampling was used to model slices of music from a dataset of Beethoven's piano sonatas. A visualization of the reduced vector space using t-distributed stochastic neighbor embedding shows that the resulting embedded vector space captures tonal relationships, even without any explicit information about the musical contents of the slices. Secondly, an excerpt of the Moonlight Sonata from Beethoven was altered by replacing slices based on context similarity. The resulting music shows that the selected slice based on similar word2vec context also has a relatively short tonal distance from the original slice. <|reference_end|>", "<|reference_start|> An ethnomusicologist contemplates universals in musical sound and musical culture: <|reference_end|>", "<|reference_start|> Convolutional Neural Networks for Sentence Classification: We report on a series of experiments with convolutional neural networks (CNN) trained on top of pre-trained word vectors for sentence-level classification tasks. We show that a simple CNN with little hyperparameter tuning and static vectors achieves excellent results on multiple benchmarks. Learning task-specific vectors through fine-tuning offers further gains in performance. We additionally propose a simple modification to the architecture to allow for the use of both task-specific and static vectors. The CNN models discussed herein improve upon the state of the art on 4 out of 7 tasks, which include sentiment analysis and question classification. <|reference_end|>" ]

[ 0, 7, 9, 11 ]

[ "<|reference_start|> Structure-From-Motion Revisited: Incremental Structure-from-Motion is a prevalent strategy for 3D reconstruction from unordered image collections. While incremental reconstruction systems have tremendously advanced in all regards, robustness, accuracy, completeness, and scalability remain the key problems towards building a truly general-purpose pipeline. We propose a new SfM technique that improves upon the state of the art to make a further step towards this ultimate goal. The full reconstruction pipeline is released to the public as an open-source implementation. <|reference_end|>", "<|reference_start|> A global linear method for camera pose registration: We present a linear method for global camera pose registration from pair wise relative poses encoded in essential matrices. Our method minimizes an approximate geometric error to enforce the triangular relationship in camera triplets. This formulation does not suffer from the typical `unbalanced scale' problem in linear methods relying on pair wise translation direction constraints, i.e. an algebraic error, nor the system degeneracy from collinear motion. In the case of three cameras, our method provides a good linear approximation of the trifocal tensor. It can be directly scaled up to register multiple cameras. The results obtained are accurate for point triangulation and can serve as a good initialization for final bundle adjustment. We evaluate the algorithm performance with different types of data and demonstrate its effectiveness. Our system produces good accuracy, robustness, and outperforms some well-known systems on efficiency. <|reference_end|>", "<|reference_start|> OpenMVG: Open Multiple View Geometry: <|reference_end|>", "<|reference_start|> Rotation Coordinate Descent for Fast Globally Optimal Rotation Averaging: Under mild conditions on the noise level of the measurements, rotation averaging satisfies strong duality, which enables global solutions to be obtained via semidefinite programming (SDP) relaxation. However, generic solvers for SDP are rather slow in practice, even on rotation averaging instances of moderate size, thus developing specialised algorithms is vital. In this paper, we present a fast algorithm that achieves global optimality called rotation coordinate descent (RCD). Unlike block coordinate descent (BCD) which solves SDP by updating the semidefinite matrix in a row-by-row fashion, RCD directly maintains and updates all valid rotations throughout the iterations. This obviates the need to store a large dense semidefinite matrix. We mathematically prove the convergence of our algorithm and empirically show its superior efficiency over state-of-the-art global methods on a variety of problem configurations. Maintaining valid rotations also facilitates incorporating local optimisation routines for further speed-ups. Moreover, our algorithm is simple to implement; see supplementary material for a demonstration program. <|reference_end|>" ]

[ 21, 47, 54, 56 ]

[ "<|reference_start|> IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2022, New Orleans, LA, USA, June 19-20, 2022: <|reference_end|>", "<|reference_start|> DeepFool: a simple and accurate method to fool deep neural networks: State-of-the-art deep neural networks have achieved impressive results on many image classification tasks. However, these same architectures have been shown to be unstable to small, well sought, perturbations of the images. Despite the importance of this phenomenon, no effective methods have been proposed to accurately compute the robustness of state-of-the-art deep classifiers to such perturbations on large-scale datasets. In this paper, we fill this gap and propose the DeepFool algorithm to efficiently compute perturbations that fool deep networks, and thus reliably quantify the robustness of these classifiers. Extensive experimental results show that our approach outperforms recent methods in the task of computing adversarial perturbations and making classifiers more robust. <|reference_end|>", "<|reference_start|> Attacking Binarized Neural Networks: Neural networks with low-precision weights and activations offer compelling efficiency advantages over their full-precision equivalents. The two most frequently discussed benefits of quantization are reduced memory consumption, and a faster forward pass when implemented with efficient bitwise operations. We propose a third benefit of very low-precision neural networks: improved robustness against some adversarial attacks, and in the worst case, performance that is on par with full-precision models. We focus on the very low-precision case where weights and activations are both quantized to $\\pm$1, and note that stochastically quantizing weights in just one layer can sharply reduce the impact of iterative attacks. We observe that non-scaled binary neural networks exhibit a similar effect to the original defensive distillation procedure that led to gradient masking, and a false notion of security. We address this by conducting both black-box and white-box experiments with binary models that do not artificially mask gradients. <|reference_end|>", "<|reference_start|> Practical Black-Box Attacks against Machine Learning: Machine learning (ML) models, e.g., deep neural networks (DNNs), are vulnerable to adversarial examples: malicious inputs modified to yield erroneous model outputs, while appearing unmodified to human observers. Potential attacks include having malicious content like malware identified as legitimate or controlling vehicle behavior. Yet, all existing adversarial example attacks require knowledge of either the model internals or its training data. We introduce the first practical demonstration of an attacker controlling a remotely hosted DNN with no such knowledge. Indeed, the only capability of our black-box adversary is to observe labels given by the DNN to chosen inputs. Our attack strategy consists in training a local model to substitute for the target DNN, using inputs synthetically generated by an adversary and labeled by the target DNN. We use the local substitute to craft adversarial examples, and find that they are misclassified by the targeted DNN. To perform a real-world and properly-blinded evaluation, we attack a DNN hosted by MetaMind, an online deep learning API. We find that their DNN misclassifies 84.24% of the adversarial examples crafted with our substitute. We demonstrate the general applicability of our strategy to many ML techniques by conducting the same attack against models hosted by Amazon and Google, using logistic regression substitutes. They yield adversarial examples misclassified by Amazon and Google at rates of 96.19% and 88.94%. We also find that this black-box attack strategy is capable of evading defense strategies previously found to make adversarial example crafting harder. <|reference_end|>" ]

[ 9, 11, 18, 22 ]

[ "<|reference_start|> Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations: We introduce a method to train Quantized Neural Networks (QNNs) --- neural networks with extremely low precision (e.g., 1-bit) weights and activations, at run-time. At train-time the quantized weights and activations are used for computing the parameter gradients. During the forward pass, QNNs drastically reduce memory size and accesses, and replace most arithmetic operations with bit-wise operations. As a result, power consumption is expected to be drastically reduced. We trained QNNs over the MNIST, CIFAR-10, SVHN and ImageNet datasets. The resulting QNNs achieve prediction accuracy comparable to their 32-bit counterparts. For example, our quantized version of AlexNet with 1-bit weights and 2-bit activations achieves $51\\%$ top-1 accuracy. Moreover, we quantize the parameter gradients to 6-bits as well which enables gradients computation using only bit-wise operation. Quantized recurrent neural networks were tested over the Penn Treebank dataset, and achieved comparable accuracy as their 32-bit counterparts using only 4-bits. Last but not least, we programmed a binary matrix multiplication GPU kernel with which it is possible to run our MNIST QNN 7 times faster than with an unoptimized GPU kernel, without suffering any loss in classification accuracy. The QNN code is available online. <|reference_end|>", "<|reference_start|> BinaryConnect: Training Deep Neural Networks with binary weights during propagations: Deep Neural Networks (DNN) have achieved state-of-the-art results in a wide range of tasks, with the best results obtained with large training sets and large models. In the past, GPUs enabled these breakthroughs because of their greater computational speed. In the future, faster computation at both training and test time is likely to be crucial for further progress and for consumer applications on low-power devices. As a result, there is much interest in research and development of dedicated hardware for Deep Learning (DL). Binary weights, i.e., weights which are constrained to only two possible values (e.g. -1 or 1), would bring great benefits to specialized DL hardware by replacing many multiply-accumulate operations by simple accumulations, as multipliers are the most space and power-hungry components of the digital implementation of neural networks. We introduce BinaryConnect, a method which consists in training a DNN with binary weights during the forward and backward propagations, while retaining precision of the stored weights in which gradients are accumulated. Like other dropout schemes, we show that BinaryConnect acts as regularizer and we obtain near state-of-the-art results with BinaryConnect on the permutation-invariant MNIST, CIFAR-10 and SVHN. <|reference_end|>", "<|reference_start|> Smoothed Analysis of Algorithms: Why the Simplex Algorithm Usually Takes Polynomial Time: We introduce the smoothed analysis of algorithms, which is a hybrid of the worst-case and average-case analysis of algorithms. In smoothed analysis, we measure the maximum over inputs of the expected performance of an algorithm under small random perturbations of that input. We measure this performance in terms of both the input size and the magnitude of the perturbations. We show that the simplex algorithm has polynomial smoothed complexity. <|reference_end|>", "<|reference_start|> Proceedings of NeurIPS 2019 Workshop on Machine Learning for the Developing World: Challenges and Risks of ML4D: This is the proceedings of the 3rd ML4D workshop which was help in Vancouver, Canada on December 13, 2019 as part of the Neural Information Processing Systems conference. <|reference_end|>" ]

[ 21, 29, 39, 41 ]

[ "<|reference_start|> Recurrent continuous translation models: We introduce a class of probabilistic continuous translation models called Recurrent Continuous Translation Models that are purely based on continuous representations for words, phrases and sentences and do not rely on alignments or phrasal translation units. The models have a generation and a conditioning aspect. The generation of the translation is modelled with a target Recurrent Language Model, whereas the conditioning on the source sentence is modelled with a Convolutional Sentence Model. Through various experiments, we show first that our models obtain a perplexity with respect to gold translations that is > 43% lower than that of stateof-the-art alignment-based translation models. Secondly, we show that they are remarkably sensitive to the word order, syntax, and meaning of the source sentence despite lacking alignments. Finally we show that they match a state-of-the-art system when rescoring n-best lists of translations. <|reference_end|>", "<|reference_start|> Convolutional Sequence to Sequence Learning: The prevalent approach to sequence to sequence learning maps an input sequence to a variable length output sequence via recurrent neural networks. We introduce an architecture based entirely on convolutional neural networks. Compared to recurrent models, computations over all elements can be fully parallelized during training and optimization is easier since the number of non-linearities is fixed and independent of the input length. Our use of gated linear units eases gradient propagation and we equip each decoder layer with a separate attention module. We outperform the accuracy of the deep LSTM setup of Wu et al. (2016) on both WMT'14 English-German and WMT'14 English-French translation at an order of magnitude faster speed, both on GPU and CPU. <|reference_end|>", "<|reference_start|> Fine-tuning for neural machine translation with limited degradation across in-and out-of-domain data: Neural machine translation is a recently proposed approach which has shown competitive results to traditional MT approaches. Similar to other neural network based methods, NMT also suffers from low performance for the domains with less available training data. Domain adaptation deals with improving performance of a model trained on large general domain data over test instances from a new domain. Fine-tuning is a fast and simple domain adaptation method which has demonstrated substantial improvements for various neural network based tasks including NMT. However, it suffers from drastic performance degradation on the general or source domain test sentences, which is undesirable in real-time applications. To address this problem of drastic degradation, in this paper, we propose two simple modifications to the fine-tuning approach, namely multi-objective learning and multi-output learning which are based on the “Knowledge distillation” framework. Experiments on English-German translations demonstrate that our approaches achieve results comparable to simple fine-tuning on the target domain task with comparatively little loss on the general domain task. <|reference_end|>", "<|reference_start|> Go From the General to the Particular: Multi-Domain Translation with Domain Transformation Networks: The key challenge of multi-domain translation lies in simultaneously encoding both the general knowledge shared across domains and the particular knowledge distinctive to each domain in a unified model. Previous work shows that the standard neural machine translation (NMT) model, trained on mixed-domain data, generally captures the general knowledge, but misses the domain-specific knowledge. In response to this problem, we augment NMT model with additional domain transformation networks to transform the general representations to domain-specific representations, which are subsequently fed to the NMT decoder. To guarantee the knowledge transformation, we also propose two complementary supervision signals by leveraging the power of knowledge distillation and adversarial learning. Experimental results on several language pairs, covering both balanced and unbalanced multi-domain translation, demonstrate the effectiveness and universality of the proposed approach. Encouragingly, the proposed unified model achieves comparable results with the fine-tuning approach that requires multiple models to preserve the particular knowledge. Further analyses reveal that the domain transformation networks successfully capture the domain-specific knowledge as expected. <|reference_end|>" ]

[ 0, 4, 13, 14 ]

[ "<|reference_start|> A Physical End-to-End Model for Molecular Communication in Nanonetworks: Molecular communication is a promising paradigm for nanoscale networks. The end-to-end (including the channel) models developed for classical wireless communication networks need to undergo a profound revision so that they can be applied for nanonetworks. Consequently, there is a need to develop new end-to-end (including the channel) models which can give new insights into the design of these nanoscale networks. The objective of this paper is to introduce a new physical end-to-end (including the channel) model for molecular communication. The new model is investigated by means of three modules, i.e., the transmitter, the signal propagation and the receiver. Each module is related to a specific process involving particle exchanges, namely, particle emission, particle diffusion and particle reception. The particle emission process involves the increase or decrease of the particle concentration rate in the environment according to a modulating input signal. The particle diffusion provides the propagation of particles from the transmitter to the receiver by means of the physics laws underlying particle diffusion in the space. The particle reception process is identified by the sensing of the particle concentration value at the receiver location. Numerical results are provided for three modules, as well as for the overall end-to-end model, in terms of normalized gain and delay as functions of the input frequency and of the transmission range. <|reference_end|>", "<|reference_start|> Diffusion-based molecular communication with limited molecule production rate: This paper studies the impact of a transmitter’s molecule generation process on the capacity of a concentration-based molecular communication (MC) system. Constraints caused by the molecule generation process affect the availability of the molecules at the transmitter. The transmitter has a storage of molecules, and should decide whether to release or save the currently produced molecules. As a result, the MC system has conceptual connections with energy harvesting systems. In this paper, we consider two scenarios on the propagation channel. The first scenario assumes a channel with no inter-symbol interference (ISI), i.e., a memoryless channel. We derive bounds on the capacity of the MC system in this scenario. The second scenario assumes an MC channel with ISI, in which the output of the channel depends on the history of released molecules in the previous time-slots. Based on the assumptions that either the transmitter or the receiver knows the channel statistics, we compute a lower bound on the channel capacity. <|reference_end|>", "<|reference_start|> Functional coupling of human pancreatic islets and liver spheroids on-a-chip: Towards a novel human ex vivo type 2 diabetes model: <|reference_end|>", "<|reference_start|> Multiscale modelling of drug transport and metabolism in liver spheroids: In early preclinical drug development, potential candidates are tested in the laboratory using isolated cells. These in vitro experiments traditionally involve cells cultured in a two-dimensional monolayer environment. However, cells cultured in three-dimensional spheroid systems have been shown to more closely resemble the functionality and morphology of cells in vivo. While the increasing usage of hepatic spheroid cultures allows for more relevant experimentation in a more realistic biological environment, the underlying physical processes of drug transport, uptake and metabolism contributing to the spatial distribution of drugs in these spheroids remain poorly understood. The development of a multiscale mathematical modelling framework describing the spatio-temporal dynamics of drugs in multicellular environments enables mechanistic insight into the behaviour of these systems. Here, our analysis of cell membrane permeation and porosity throughout the spheroid reveals the impact of these properties on drug penetration, with maximal disparity between zonal metabolism rates occurring for drugs of intermediate lipophilicity. Our research shows how mathematical models can be used to simulate the activity and transport of drugs in hepatic spheroids and in principle any organoid, with the ultimate aim of better informing experimentalists on how to regulate dosing and culture conditions to more effectively optimize drug delivery. <|reference_end|>" ]

[ 4, 7, 20, 22 ]

[ "<|reference_start|> Why researchers should think \"within-person\": A paradigmatic rationale.: <|reference_end|>", "<|reference_start|> The New Person-Specific Paradigm in Psychology: Most research methodology in the behavioral sciences employs interindividual analyses, which provide information about the state of affairs of the population. However, as shown by classical mathematical-statistical theorems (the ergodic theorems), such analyses do not provide information for, and cannot be applied at, the level of the individual, except on rare occasions when the processes of interest meet certain stringent conditions. When psychological processes violate these conditions, the interindividual analyses that are now standardly applied have to be replaced by analysis of intraindividual variation in order to obtain valid results. Two illustrations involving analysis of intraindividual variation of personality and emotional processes are given. <|reference_end|>", "<|reference_start|> Ecological Momentary Assessment: A New Tool for Behavioral Medicine Research: <|reference_end|>", "<|reference_start|> Macroeconomics and Reality: Existing strategies for econometric analysis related to macroeconomics are subject to a number of serious objections, some recently formulated, some old. These objections are summarized in this paper, and it is argued that taken together they make it unlikely that macroeconomic models are in fact over identified, as the existing statistical theory usually assumes. The implications of this conclusion are explored, and an example of econometric work in a non-standard style, taking account of the objections to the standard style, is presented. THE STUDY OF THE BUSINESS cycle, fluctuations in aggregate measures of economic activity and prices over periods from one to ten years or so, constitutes or motivates a large part of what we call macroeconomics. Most economists would agree that there are many macroeconomic variables whose cyclical fluctuations are of interest, and would agree further that fluctuations in these series are interrelated. It would seem to follow almost tautologically that statistical models involving large numbers of macroeconomic variables ought to be the arena within which macroeconomic theories confront reality and thereby each other. Instead, though large-scale statistical macroeconomic models exist and are by some criteria successful, a deep vein of skepticism about the value of these models runs through that part of the economics profession not actively engaged in constructing or using them. It is still rare for empirical research in macroeconomics to be planned and executed within the framework of one of the large models. In this lecture I intend to discuss some aspects of this situation, attempting both to offer some explanations and to suggest some means for improvement. I will argue that the style in which their builders construct claims for a connection between these models and reality-the style in which \"identification\" is achieved for these models-is inappropriate, to the point at which claims for identification in these models cannot be taken seriously. This is a venerable assertion; and there are some good old reasons for believing it;2 but there are also some reasons which have been more recently put forth. After developing the conclusion that the identification claimed for existing large-scale models is incredible, I will discuss what ought to be done in consequence. The line of argument is: large-scale models do perform useful forecasting and policy-analysis functions despite their incredible identification; the restrictions imposed in the usual style of identification are neither essential to constructing a model which can perform these functions nor innocuous; an alternative style of identification is available and practical. Finally we will look at some empirical work based on an alternative style of macroeconometrics. A six-variable dynamic system is estimated without using 1 Research for this paper was supported by NSF Grant Soc-76-02482. Lars Hansen executed the computations. The paper has benefited from comments by many people, especially Thomas J. Sargent <|reference_end|>" ]

[ 4, 7, 12, 13 ]

[ "<|reference_start|> Piecewise Normalizing Flows: Normalizing flows are an established approach for modelling complex probability densities through invertible transformations from a base distribution. However, the accuracy with which the target distribution can be captured by the normalizing flow is strongly influenced by the topology of the base distribution. A mismatch between the topology of the target and the base can result in a poor performance, as is typically the case for multi-modal problems. A number of different works have attempted to modify the topology of the base distribution to better match the target, either through the use of Gaussian Mixture Models (Izmailov et al., 2020; Ardizzone et al., 2020; Hagemann&Neumayer, 2021) or learned accept/reject sampling (Stimper et al., 2022). We introduce piecewise normalizing flows which divide the target distribution into clusters, with topologies that better match the standard normal base distribution, and train a series of flows to model complex multi-modal targets. We demonstrate the performance of the piecewise flows using some standard benchmarks and compare the accuracy of the flows to the approach taken in Stimper et al. (2022) for modelling multi-modal distributions. We find that our approach consistently outperforms the approach in Stimper et al. (2022) with a higher emulation accuracy on the standard benchmarks. <|reference_end|>", "<|reference_start|> Analyzing and improving a wireless network: This report is about how a site survey is carried out using software and how a site survey can collect the information needed to improve the existing wireless networks. The report provides both general information about wireless networks and practical examples with authentic measurements in the corporate environment.Problems can arise in a wireless network if the planning is not properly done. A measurement have been made in second floor, where company X is located.The results of the survey is presented in images that show the strength of Company X wireless networks, on the second floor. Images also shows noise, overlapping channels and how the signal strength related to these. <|reference_end|>", "<|reference_start|> Statistical inference: What is Statistics? Opinions vary. In fact, there is a continuous spectrum of attitudes toward statistics ranging from pure theoreticians, proving asymptotic efficiency and searching for most powerful tests, to wild practitioners, blindly reporting p-values and claiming statistical significance for scientifically insignificant results. In these notes statistics is viewed as a branch of mathematical engineering, that studies ways of extracting reliable information from limited data for learning, prediction, and decision making in the presence of uncertainty. The main goals of these notes are: (1) provide a logical introduction to statistical inference, (2) develop statistical thinking and intuitive feel for the subject, and (3) introduce the most fundamental ideas, concepts, and methods of statistics, explain how and why they work, and when they don’t. These lecture notes are based on the courses the author taught at the University of Southern California in 2012 and 2013, and at the California Institute of Technology in 2016 and 2017. <|reference_end|>", "<|reference_start|> Diagnosing and Enhancing VAE Models: Although variational autoencoders (VAEs) represent a widely influential deep generative model, many aspects of the underlying energy function remain poorly understood. In particular, it is commonly believed that Gaussian encoder/decoder assumptions reduce the effectiveness of VAEs in generating realistic samples. In this regard, we rigorously analyze the VAE objective, differentiating situations where this belief is and is not actually true. We then leverage the corresponding insights to develop a simple VAE enhancement that requires no additional hyperparameters or sensitive tuning. Quantitatively, this proposal produces crisp samples and stable FID scores that are actually competitive with a variety of GAN models, all while retaining desirable attributes of the original VAE architecture. A shorter version of this work will appear in the ICLR 2019 conference proceedings (Dai and Wipf, 2019). The code for our model is available at this https URL TwoStageVAE. <|reference_end|>" ]

[ 1, 5, 10, 18 ]

[ "<|reference_start|> Personalized Decision Making -- A Conceptual Introduction: Personalized decision making targets the behavior of a specific individual, while population-based decision making concerns a sub-population resembling that individual. This paper clarifies the distinction between the two and explains why the former leads to more informed decisions. We further show that by combining experimental and observational studies we can obtain valuable information about individual behavior and, consequently, improve decisions over those obtained from experimental studies alone. <|reference_end|>", "<|reference_start|> Probabilities Of Causation: Three Counterfactual Interpretations And Their Identification: <|reference_end|>", "<|reference_start|> An Axiomatic Characterization of Causal Counterfactuals: <|reference_end|>", "<|reference_start|> Bounds on Causal Effects and Application to High Dimensional Data: This paper addresses the problem of estimating causal effects when adjustment variables in the back-door or front-door criterion are partially observed. For such scenarios, we derive bounds on the causal effects by solving two non-linear optimization problems, and demonstrate that the bounds are sufficient. Using this optimization method, we propose a framework for dimensionality reduction that allows one to trade bias for estimation power, and demonstrate its performance using simulation studies. <|reference_end|>" ]

[ 1, 2, 3, 12 ]

[ "<|reference_start|> Tight consistency bounds for bitcoin: We establish the optimal security threshold for the Bitcoin protocol in terms of adversarial hashing power, honest hashing power, and network delays. Specifically, we prove that the protocol is secure if [ra < 1/Δ0 + 1/rh,,] where rh is the expected number of honest proof-of-work successes in unit time, ra is the expected number of adversarial successes, and no message is delayed by more than Δ0 time units. In this regime, the protocol guarantees consistency and liveness with exponentially decaying failure probabilities. Outside this region, the simple private chain attack prevents consensus. Our analysis immediately applies to any Nakamoto-style proof-of-work protocol; in the full version of this paper we also present the adaptations needed to apply it in the proof-of-stake setting, establishing a similar threshold there. <|reference_end|>", "<|reference_start|> How Does Nakamoto Set His Clock? Full Analysis of Nakamoto Consensus in Bounded Delay Networks: . Nakamoto consensus, arguably the most exciting development in distributed computing in the last few years, is in a sense a recasting of the traditional state-machine-replication problem in an unauthenticated setting, where furthermore parties come and go without warning. The protocol relies on a cryptographic primitive known as proof of work (PoW) which is used to throttle message passing. Importantly, the PoW difficulty level is appropriately adjusted throughout the course of the protocol execution relying on the blockchain’s timekeeping ability. While the original formulation was only accompanied by rudimentary analysis, significant and steady progress has been made in abstracting the protocol’s properties and providing a formal analysis under various restrictions and protocol simplifications. Still, a full analysis of the protocol that includes its target recalculation and, notably, the timestamp adjustment mechanism —specifically, the protocol allows incoming block timestamps in the near future, as determined by a protocol parameter, and rejects blocks that have a timestamp in the past of the median time of a specific number of blocks on-chain (namely, 11)— which equip it to operate in its intended setting of bounded communication delays, imperfect clocks and dynamic participation, has remained open. The gap is that Nakamoto’s protocol fundamentally depends on the blockchain itself to be a consistent timekeeper that should advance roughly on par with real time. In order to tackle this question we introduce a new analytical tool that we call hot-hand executions , which capture the regular occurrence of high concentration of honestly generated blocks, and correspondingly put forth and prove a new blockchain property called concentrated chain quality , which may be of independent interest. Utilizing these tools and techniques we demonstrate that Nakamoto’s protocol achieves, under suitable conditions, safety, liveness as well as (consistent) timekeeping. <|reference_end|>", "<|reference_start|> Ebb-and-Flow Protocols: A Resolution of the Availability-Finality Dilemma: The CAP theorem says that no blockchain can be live under dynamic participation and safe under temporary network partitions. To resolve this availability-finality dilemma, we formulate a new class of flexible consensus protocols, ebb-and-flow protocols, which support a full dynamically available ledger in conjunction with a finalized prefix ledger. The finalized ledger falls behind the full ledger when the network partitions but catches up when the network heals. Gasper, the current candidate protocol for Ethereum 2.0's beacon chain, combines the finality gadget Casper FFG with the LMD GHOST fork choice rule and aims to achieve this property. However, we discovered an attack in the standard synchronous network model, highlighting a general difficulty with existing finality-gadget-based designs. We present a construction of provably secure ebb-and-flow protocols with optimal resilience. Nodes run an off-the-shelf dynamically available protocol, take snapshots of the growing available ledger, and input them into a separate off-the-shelf BFT protocol to finalize a prefix. We explore connections with flexible BFT and improve upon the state-of-the-art for that problem. <|reference_end|>", "<|reference_start|> The Longest-Chain Protocol Under Random Delays: In the field of distributed consensus and blockchains, the synchronous communication model assumes that all messages between honest parties are delayed at most by a known constant $\\Delta$. Recent literature establishes that the longest-chain blockchain protocol is secure under the synchronous model. However, for a fixed mining rate, the security guarantees degrade with $\\Delta$. We analyze the performance of the longest-chain protocol under the assumption that the communication delays are random, independent, and identically distributed. This communication model allows for distributions with unbounded support and is a strict generalization of the synchronous model. We provide safety and liveness guarantees with simple, explicit bounds on the failure probabilities. These bounds hold for infinite-horizon executions and decay exponentially with the security parameter. In particular, we show that the longest-chain protocol has good security guarantees when delays are sporadically large and possibly unbounded, which is reflective of real-world network conditions. <|reference_end|>" ]

[ 2, 8, 15, 20 ]

[ "<|reference_start|> A New Performance Measure and Evaluation Benchmark for Road Detection Algorithms: Detecting the road area and ego-lane ahead of a vehicle is central to modern driver assistance systems. While lane-detection on well-marked roads is already available in modern vehicles, finding the boundaries of unmarked or weakly marked roads and lanes as they appear in inner-city and rural environments remains an unsolved problem due to the high variability in scene layout and illumination conditions, amongst others. While recent years have witnessed great interest in this subject, to date no commonly agreed upon benchmark exists, rendering a fair comparison amongst methods difficult. In this paper, we introduce a novel open-access dataset and benchmark for road area and ego-lane detection. Our dataset comprises 600 annotated training and test images of high variability from the KITTI autonomous driving project, capturing a broad spectrum of urban road scenes. For evaluation, we propose to use the 2D Bird's Eye View (BEV) space as vehicle control usually happens in this 2D world, requiring detection results to be represented in this very same space. Furthermore, we propose a novel, behavior-based metric which judges the utility of the extracted ego-lane area for driver assistance applications by fitting a driving corridor to the road detection results in the BEV. We believe this to be important for a meaningful evaluation as pixel-level performance is of limited value for vehicle control. State-of-the-art road detection algorithms are used to demonstrate results using classical pixel-level metrics in perspective and BEV space as well as the novel behavior-based performance measure. All data and annotations are made publicly available on the KITTI online evaluation website in order to serve as a common benchmark for road terrain detection algorithms. <|reference_end|>", "<|reference_start|> Modeling the space of camera response functions: Many vision applications require precise measurement of scene radiance. The function relating scene radiance to image intensity of an imaging system is called the camera response. We analyze the properties that all camera responses share. This allows us to find the constraints that any response function must satisfy. These constraints determine the theoretical space of all possible camera responses. We have collected a diverse database of real-world camera response functions (DoRF). Using this database, we show that real-world responses occupy a small part of the theoretical space of all possible responses. We combine the constraints from our theoretical space with the data from DoRF to create a low-parameter empirical model of response (EMoR). This response model allows us to accurately interpolate the complete response function of a camera from a small number of measurements obtained using a standard chart. We also show that the model can be used to accurately estimate the camera response from images of an arbitrary scene taken using different exposures. The DoRF database and the EMoR model can be downloaded at http://www.cs.columbia.edu/CAVE. <|reference_end|>", "<|reference_start|> {Learning to estimate and remove non-uniform image blur: This paper addresses the problem of restoring images subjected to unknown and spatially varying blur caused by defocus or linear (say, horizontal) motion. The estimation of the global (non-uniform) image blur is cast as a multi-label energy minimization problem. The energy is the sum of unary terms corresponding to learned local blur estimators, and binary ones corresponding to blur smoothness. Its global minimum is found using Ishikawa's method by exploiting the natural order of discretized blur values for linear motions and defocus. Once the blur has been estimated, the image is restored using a robust (non-uniform) deblurring algorithm based on sparse regularization with global image statistics. The proposed algorithm outputs both a segmentation of the image into uniform-blur layers and an estimate of the corresponding sharp image. We present qualitative results on real images, and use synthetic data to quantitatively compare our approach to the publicly available implementation of Chakrabarti~et al. <|reference_end|>", "<|reference_start|> SUN RGB-D: A rgb-d scene understanding benchmark suite: Although RGB-D sensors have enabled major break-throughs for several vision tasks, such as 3D reconstruction, we have not attained the same level of success in high-level scene understanding. Perhaps one of the main reasons is the lack of a large-scale benchmark with 3D annotations and 3D evaluation metrics. In this paper, we introduce an RGB-D benchmark suite for the goal of advancing the state-of-the-arts in all major scene understanding tasks. Our dataset is captured by four different sensors and contains 10,335 RGB-D images, at a similar scale as PASCAL VOC. The whole dataset is densely annotated and includes 146,617 2D polygons and 64,595 3D bounding boxes with accurate object orientations, as well as a 3D room layout and scene category for each image. This dataset enables us to train data-hungry algorithms for scene-understanding tasks, evaluate them using meaningful 3D metrics, avoid overfitting to a small testing set, and study cross-sensor bias. <|reference_end|>" ]

[ 1, 10, 11, 14 ]

[ "<|reference_start|> Evaluation of uav photogrammetric accuracy for mapping and earthworks computations: This study quantifies the accuracies achieved and tests the validity of an in-house developed Unmanned Aerial Vehicle (UAV) system employed in a stockpile volumetric survey. UAV photogrammetric results are compared with conventional GNSS survey results. To test the repeatability of the UAV system, multiple flights were flown over the same stockpile using different GNSS ground control, at different times and weather conditions. Positional accuracies of UAV photogrammetric results were found to be very similar to those from GNSS RTK survey, at the scale of photography flown. UAV stockpile volume results agreed with those from GNSS within 3 755 m3 (0.7%) on a 530 255 m3 pile. Stockpile volume comparisons between subsequent UAV surface models agreed within 877 m3 (0.2%) on the same pile. Geometric analysis of independent UAV photogrammetric models over the same area indicated that they could be considered the same at a 95% confidence level. We conclude that the UAV photogrammetric approach is, at the very lea... <|reference_end|>", "<|reference_start|> Photogrammetric exploitation of ikonos imagery for mapping applications: The launch of IKONOS by Space Imaging opens a new era of high-resolution satellite imagery collection and mapping. The IKONOS satellite simultaneously acquires 1 m panchromatic and 4 m multi-spectral images in four bands that are suitable for high accuracy mapping applications. Space Imaging uses the rational function model (RFM), also known as rational polynomial camera model, instead of the physical IKONOS sensor model to communicate the imaging geometry. As revealed by recent studies from several researchers, the RFM retains the full capability of performing photogrammetric processing in absence of the physical sensor model. This paper presents some RFM-based processing methods and mapping applications developed for 3D feature extraction, orthorectification and RPC model refinement using IKONOS imagery. Comprehensive tests are performed to test the accuracy of 3D reconstruction and orthorectification and to validate the feasibility of the model refinement techniques. <|reference_end|>", "<|reference_start|> Automatic car counting method for unmanned aerial vehicle images: This paper presents a solution to solve the car detection and counting problem in images acquired by means of unmanned aerial vehicles (UAVs). UAV images are characterized by a very high spatial resolution (order of few centimeters), and consequently by an extremely high level of details which calls for appropriate automatic analysis methods. The proposed method starts with a screening step of asphalted zones in order to restrict the areas where to detect cars and thus to reduce false alarms. Then, it performs a feature extraction process based on scalar invariant feature transform thanks to which a set of keypoints is identified in the considered image and opportunely described. Successively, it discriminates between keypoints assigned to cars and all the others, by means of a support vector machine classifier. The last step of our method is focused on the grouping of the keypoints belonging to the same car in order to get a “one keypoint-one car” relationship. Finally, the number of cars present in the scene is given by the number of final keypoints identified. The experimental results obtained on a real UAV scene characterized by a spatial resolution of 2 cm show that the proposed method exhibits a promising car counting accuracy. <|reference_end|>", "<|reference_start|> Semi-automatic detection and counting of oil palm trees from high spatial resolution airborne imagery: Plantation inventory and management require a range of fine-scale remote-sensing data. Remote-sensing images with high spatial and spectral resolution are an efficient source of such information. This article presents an approach to the extraction and counting of oil palm trees from high spatial resolution airborne imagery data. Counting oil palm trees is a crucial problem in specific agricultural areas, especially in Malaysia. The proposed scheme comprises six major parts: (1) discrimination of oil palms from non-oil palms using spectral analysis, (2) texture analysis, (3) edge enhancement, (4) segmentation process, (5) morphological analysis and (6) blob analysis. The average accuracy obtained was 95%, which indicates that high spatial resolution airborne imagery data with an appropriate assessment technique have the potential to provide us with vital information for oil palm plantation management. Information on the number of oil palm trees is crucial to the ability of plantation management to assess the value of the plantation and to monitor its production. <|reference_end|>" ]

[ 1, 3, 4, 7 ]

[ "<|reference_start|> Schur Number Five: We present the solution of a century-old problem known as Schur Number Five: What is the largest (natural) number $n$ such that there exists a five-coloring of the positive numbers up to $n$ without a monochromatic solution of the equation $a + b = c$? We obtained the solution, $n = 160$, by encoding the problem into propositional logic and applying massively parallel satisfiability solving techniques on the resulting formula. We constructed and validated a proof of the solution to increase trust in the correctness of the multi-CPU-year computations. The proof is two petabytes in size and was certified using a formally verified proof checker, demonstrating that any result by satisfiability solvers---no matter how large---can now be validated using highly trustworthy systems. <|reference_end|>", "<|reference_start|> Inprocessing Rules: <|reference_end|>", "<|reference_start|> Many hard examples for resolution: For every choice of positive integers <italic>c</italic> and <italic>k</italic> such that <italic>k</italic> ≥ 3 and <italic>c</italic>2<supscrpt>-<italic>k</italic></supscrpt> ≥ 0.7, there is a positive number ε such that, with probability tending to 1 as <italic>n</italic> tends to ∞, a randomly chosen family of <italic>cn</italic> clauses of size <italic>k</italic> over <italic>n</italic> variables is unsatisfiable, but every resolution proof of its unsatisfiability must generate at least (1 + ε)<supscrpt><italic>n</italic></supscrpt> clauses. <|reference_end|>", "<|reference_start|> Inprocessing Rules: <|reference_end|>" ]

[ 1, 8, 21, 33 ]

[ "<|reference_start|> Playing Atari with Deep Reinforcement Learning: We present the first deep learning model to successfully learn control policies directly from high-dimensional sensory input using reinforcement learning. The model is a convolutional neural network, trained with a variant of Q-learning, whose input is raw pixels and whose output is a value function estimating future rewards. We apply our method to seven Atari 2600 games from the Arcade Learning Environment, with no adjustment of the architecture or learning algorithm. We find that it outperforms all previous approaches on six of the games and surpasses a human expert on three of them. <|reference_end|>", "<|reference_start|> Continuous control with deep reinforcement learning: We adapt the ideas underlying the success of Deep Q-Learning to the continuous action domain. We present an actor-critic, model-free algorithm based on the deterministic policy gradient that can operate over continuous action spaces. Using the same learning algorithm, network architecture and hyper-parameters, our algorithm robustly solves more than 20 simulated physics tasks, including classic problems such as cartpole swing-up, dexterous manipulation, legged locomotion and car driving. Our algorithm is able to find policies whose performance is competitive with those found by a planning algorithm with full access to the dynamics of the domain and its derivatives. We further demonstrate that for many of the tasks the algorithm can learn policies end-to-end: directly from raw pixel inputs. <|reference_end|>", "<|reference_start|> Policy Gradient for Continuing Tasks in Non-stationary Markov Decision Processes: Reinforcement learning considers the problem of finding policies that maximize an expected cumulative reward in a Markov decision process with unknown transition probabilities. In this paper we consider the problem of finding optimal policies assuming that they belong to a reproducing kernel Hilbert space (RKHS). To that end we compute unbiased stochastic gradients of the value function which we use as ascent directions to update the policy. A major drawback of policy gradient-type algorithms is that they are limited to episodic tasks unless stationarity assumptions are imposed. Hence preventing these algorithms to be fully implemented online, which is a desirable property for systems that need to adapt to new tasks and/or environments in deployment. The main requirement for a policy gradient algorithm to work is that the estimate of the gradient at any point in time is an ascent direction for the initial value function. In this work we establish that indeed this is the case which enables to show the convergence of the online algorithm to the critical points of the initial value function. A numerical example shows the ability of our online algorithm to learn to solve a navigation and surveillance problem, in which an agent must loop between to goal locations. This example corroborates our theoretical findings about the ascent directions of subsequent stochastic gradients. It also shows how the agent running our online algorithm succeeds in learning to navigate, following a continuing cyclic trajectory that does not comply with the standard stationarity assumptions in the literature for non episodic training. <|reference_end|>", "<|reference_start|> Deep reinforcement learning amidst continual structured non-stationarity: As humans, our goals and our environment are persistently changing throughout our lifetime based on our experiences, actions, and internal and external drives. In contrast, typical reinforcement learning problem set-ups consider decision processes that are stationary across episodes. Can we develop reinforcement learning algorithms that can cope with the persistent change in the former, more realistic problem settings? While on-policy algorithms such as policy gradients in principle can be extended to non-stationary settings, the same cannot be said for more efficient off-policy algorithms that replay past experiences when learning. In this work, we formalize this problem setting, and draw upon ideas from the online learning and probabilistic inference literature to derive an off-policy RL algorithm that can reason about and tackle such lifelong non-stationarity. Our method leverages latent variable models to learn a representation of the environment from current and past experiences, and performs off-policy RL with this representation. We further introduce several simulation environments that exhibit lifelong non-stationarity, and empirically find that our approach substantially outperforms approaches that do not reason about environment shift. <|reference_end|>" ]

[ 0, 2, 9, 20 ]

[ "<|reference_start|> {PAC-Bayesian Generalisation Error Bounds for Gaussian Process Classification: Approximate Bayesian Gaussian process (GP) classification techniques are powerful non-parametric learning methods, similar in appearance and performance to support vector machines. Based on simple probabilistic models, they render interpretable results and can be embedded in Bayesian frameworks for model selection, feature selection, etc. In this paper, by applying the PAC-Bayesian theorem of McAllester (1999a), we prove distribution-free generalisation error bounds for a wide range of approximate Bayesian GP classification techniques. We also provide a new and much simplified proof for this powerful theorem, making use of the concept of convex duality which is a backbone of many machine learning techniques. We instantiate and test our bounds for two particular GPC techniques, including a recent sparse method which circumvents the unfavourable scaling of standard GP algorithms. As is shown in experiments on a real-world task, the bounds can be very tight for moderate training sample sizes. To the best of our knowledge, these results provide the tightest known distribution-free error bounds for approximate Bayesian GPC methods, giving a strong learning-theoretical justification for the use of these techniques. <|reference_end|>", "<|reference_start|> Neural Ordinary Differential Equations: We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver. These continuous-depth models have constant memory cost, adapt their evaluation strategy to each input, and can explicitly trade numerical precision for speed. We demonstrate these properties in continuous-depth residual networks and continuous-time latent variable models. We also construct continuous normalizing flows, a generative model that can train by maximum likelihood, without partitioning or ordering the data dimensions. For training, we show how to scalably backpropagate through any ODE solver, without access to its internal operations. This allows end-to-end training of ODEs within larger models. <|reference_end|>", "<|reference_start|> ODE$^2$VAE: Deep generative second order ODEs with Bayesian neural networks: We present Ordinary Differential Equation Variational Auto-Encoder (ODE$^2$VAE), a latent second order ODE model for high-dimensional sequential data. Leveraging the advances in deep generative models, ODE$^2$VAE can simultaneously learn the embedding of high dimensional trajectories and infer arbitrarily complex continuous-time latent dynamics. Our model explicitly decomposes the latent space into momentum and position components and solves a second order ODE system, which is in contrast to recurrent neural network (RNN) based time series models and recently proposed black-box ODE techniques. In order to account for uncertainty, we propose probabilistic latent ODE dynamics parameterized by deep Bayesian neural networks. We demonstrate our approach on motion capture, image rotation and bouncing balls datasets. We achieve state-of-the-art performance in long term motion prediction and imputation tasks. <|reference_end|>", "<|reference_start|> Differential Bayesian Neural Nets: Neural Ordinary Differential Equations (N-ODEs) are a powerful building block for learning systems, which extend residual networks to a continuous-time dynamical system. We propose a Bayesian version of N-ODEs that enables well-calibrated quantification of prediction uncertainty, while maintaining the expressive power of their deterministic counterpart. We assign Bayesian Neural Nets (BNNs) to both the drift and the diffusion terms of a Stochastic Differential Equation (SDE) that models the flow of the activation map in time. We infer the posterior on the BNN weights using a straightforward adaptation of Stochastic Gradient Langevin Dynamics (SGLD). We illustrate significantly improved stability on two synthetic time series prediction tasks and report better model fit on UCI regression benchmarks with our method when compared to its non-Bayesian counterpart. <|reference_end|>" ]

[ 2, 4, 7, 14 ]

[ "<|reference_start|> {A Family of Algorithms for Approximate Bayesian Inference: One of the major obstacles to using Bayesian methods for pattern recognition has been its computational expense. This thesis presents an approximation technique that can perform Bayesian inference faster and more accurately than previously possible. This method, “Expectation Propagation,” unifies and generalizes two previous techniques: assumed-density filtering, an extension of the Kalman filter, and loopy belief propagation, an extension of belief propagation in Bayesian networks. The unification shows how both of these algorithms can be viewed as approximating the true posterior distribution with simpler distribution, which is close in the sense of KL-divergence. Expectation Propagation exploits the best of both algorithms: the generality of assumed-density filtering and the accuracy of loopy belief propagation. \nLoopy belief propagation, because it propagates exact belief states, is useful for limited types of belief networks, such as purely discrete networks. Expectation Propagation approximates the belief states with expectations, such as means and variances, giving it much wider scope. Expectation Propagation also extends belief propagation in the opposite direction—propagating richer belief states which incorporate correlations between variables. \nThis framework is demonstrated in a variety of statistical models using synthetic and real-world data. On Gaussian mixture problems, Expectation Propagation is found, for the same amount of computation, to be convincingly better than rival approximation techniques: Monte Carlo, Laplace's method, and variational Bayes. For pattern recognition, Expectation Propagation provides an algorithm for training Bayes Point Machine classifiers that is faster and more accurate than any previously known. The resulting classifiers outperform Support Vector Machines on several standard datasets, in addition to having a comparable training time. Expectation Propagation can also be used to choose an appropriate feature set for classification, via Bayesian model selection. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.) <|reference_end|>", "<|reference_start|> Graphical models, Exponential families, and variational inference: The formalism of probabilistic graphical models provides a unifying framework for capturing complex dependencies among random variables, and building large-scale multivariate statistical models. Graphical models have become a focus of research in many statistical, computational and mathematical fields, including bioinformatics, communication theory, statistical physics, combinatorial optimization, signal and image processing, information retrieval and statistical machine learning. Many problems that arise in specific instances — including the key problems of computing marginals and modes of probability distributions — are best studied in the general setting. Working with exponential family representations, and exploiting the conjugate duality between the cumulant function and the entropy for exponential families, we develop general variational representations of the problems of computing likelihoods, marginal probabilities and most probable configurations. We describe how a wide variety of algorithms — among them sum-product, cluster variational methods, expectation-propagation, mean field methods, max-product and linear programming relaxation, as well as conic programming relaxations — can all be understood in terms of exact or approximate forms of these variational representations. The variational approach provides a complementary alternative to Markov chain Monte Carlo as a general source of approximation methods for inference in large-scale statistical models. <|reference_end|>", "<|reference_start|> {A Family of Algorithms for Approximate Bayesian Inference: One of the major obstacles to using Bayesian methods for pattern recognition has been its computational expense. This thesis presents an approximation technique that can perform Bayesian inference faster and more accurately than previously possible. This method, “Expectation Propagation,” unifies and generalizes two previous techniques: assumed-density filtering, an extension of the Kalman filter, and loopy belief propagation, an extension of belief propagation in Bayesian networks. The unification shows how both of these algorithms can be viewed as approximating the true posterior distribution with simpler distribution, which is close in the sense of KL-divergence. Expectation Propagation exploits the best of both algorithms: the generality of assumed-density filtering and the accuracy of loopy belief propagation. \nLoopy belief propagation, because it propagates exact belief states, is useful for limited types of belief networks, such as purely discrete networks. Expectation Propagation approximates the belief states with expectations, such as means and variances, giving it much wider scope. Expectation Propagation also extends belief propagation in the opposite direction—propagating richer belief states which incorporate correlations between variables. \nThis framework is demonstrated in a variety of statistical models using synthetic and real-world data. On Gaussian mixture problems, Expectation Propagation is found, for the same amount of computation, to be convincingly better than rival approximation techniques: Monte Carlo, Laplace's method, and variational Bayes. For pattern recognition, Expectation Propagation provides an algorithm for training Bayes Point Machine classifiers that is faster and more accurate than any previously known. The resulting classifiers outperform Support Vector Machines on several standard datasets, in addition to having a comparable training time. Expectation Propagation can also be used to choose an appropriate feature set for classification, via Bayesian model selection. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.) <|reference_end|>", "<|reference_start|> {Approximate inference techniques with expectation constraints: This paper discusses inference problems in probabilistic graphical models that often occur in a machine learning setting. In particular it presents a unified view of several recently proposed approximation schemes. Expectation consistent approximations and expectation propagation are both shown to be related to Bethe free energies with weak consistency constraints, i.e. free energies where local approximations are only required to agree on certain statistics instead of full marginals. <|reference_end|>" ]

[ 1, 3, 8, 9 ]

[ "<|reference_start|> Improving PILCO with bayesian neural network dynamics models: Model-based reinforcement learning (RL) allows an agent to discover good policies with a small number of trials by generalising observed transitions. Data efficiency can be further improved with a probabilistic model of the agent’s ignorance about the world, allowing it to choose actions under uncertainty. Bayesian modelling offers tools for this task, with PILCO [1] being a prominent example, achieving state-of-theart data efficiency on low dimensional RL benchmarks. But PILCO relies on Gaussian processes (GPs), which prohibits its applicability to problems that require a larger number of trials to be solved. Further, PILCO does not consider temporal correlation in model uncertainty between successive state transitions, which results in PILCO underestimating state uncertainty at future time steps [2]. In this paper we extend PILCO’s framework to use Bayesian deep dynamics models with approximate variational inference, allowing PILCO to scale linearly with number of trials and observation space dimensionality. Using particle methods we sample dynamics function realisations, and obtain lower cumulative cost than PILCO. We give insights into the modelling assumptions made in PILCO, and show that moment matching is a crucial simplifying assumption made by the model. Our implementation can leverage GPU architectures, offering faster running time than PILCO, and will allow structured observation spaces to be modelled (images or higher dimensional inputs) in the future. <|reference_end|>", "<|reference_start|> Learning From Demonstration: By now it is widely accepted that learning a task from scratch, i.e., without any prior knowledge, is a daunting undertaking. Humans, however, rarely attempt to learn from scratch. They extract initial biases as well as strategies how to approach a learning problem from instructions and/or demonstrations of other humans. For teaming control, this paper investigates how learning from demonstration can be applied in the context of reinforcement learning. We consider priming the Q-function, the value function, the policy, and the model of the task dynamics as possible areas where demonstrations can speed up learning. In general nonlinear learning problems, only model-based reinforcement learning shows significant speed-up after a demonstration, while in the special case of linear quadratic regulator (LQR) problems, all methods profit from the demonstration. In an implementation of pole balancing on a complex anthropomorphic robot arm, we demonstrate that, when facing the complexities of real signal processing, model-based reinforcement learning offers the most robustness for LQR problems. Using the suggested methods, the robot learns pole balancing in just a single trial after a 30 second long demonstration of the human instructor. <|reference_end|>", "<|reference_start|> Task Transfer by Preference-Based Cost Learning: The goal of task transfer in reinforcement learning is migrating the action policy of an agent to the target task from the source task. Given their successes on robotic action planning, current methods mostly rely on two requirements: exactly-relevant expert demonstrations or the explicitly-coded cost function on target task, both of which, however, are inconvenient to obtain in practice. In this paper, we relax these two strong conditions by developing a novel task transfer framework where the expert preference is applied as a guidance. In particular, we alternate the following two steps: Firstly, letting experts apply pre-defined preference rules to select related expert demonstrates for the target task. Secondly, based on the selection result, we learn the target cost function and trajectory distribution simultaneously via enhanced Adversarial MaxEnt IRL and generate more trajectories by the learned target distribution for the next preference selection. The theoretical analysis on the distribution learning and convergence of the proposed algorithm are provided. Extensive simulations on several benchmarks have been conducted for further verifying the effectiveness of the proposed method. <|reference_end|>", "<|reference_start|> Large deflection dynamics and control for planar continuum robots: This paper focuses on a class of robot manipulators termed \"continuum\" robots - robots that exhibit behavior similar to tentacles, trunks, and snakes. In previous work, we studied details of the mechanical design, kinematics, path-planning and small-deflection dynamics for continuum robots such as the Clemson \"tentacle manipulator\". In this paper, we discuss the dynamics of a planar continuum backbone section, incorporating a large-deflection dynamic model. Based on these dynamics, we formulate a vibration-damping setpoint controller, and include experimental results to illustrate the efficacy of the proposed controller. <|reference_end|>" ]

[ 13, 21, 23, 37 ]

[ "<|reference_start|> Pseudo Numerical Methods for Diffusion Models on Manifolds: Denoising Diffusion Probabilistic Models (DDPMs) can generate high-quality samples such as image and audio samples. However, DDPMs require hundreds to thousands of iterations to produce final samples. Several prior works have successfully accelerated DDPMs through adjusting the variance schedule (e.g., Improved Denoising Diffusion Probabilistic Models) or the denoising equation (e.g., Denoising Diffusion Implicit Models (DDIMs)). However, these acceleration methods cannot maintain the quality of samples and even introduce new noise at a high speedup rate, which limit their practicability. To accelerate the inference process while keeping the sample quality, we provide a fresh perspective that DDPMs should be treated as solving differential equations on manifolds. Under such a perspective, we propose pseudo numerical methods for diffusion models (PNDMs). Specifically, we figure out how to solve differential equations on manifolds and show that DDIMs are simple cases of pseudo numerical methods. We change several classical numerical methods to corresponding pseudo numerical methods and find that the pseudo linear multi-step method is the best in most situations. According to our experiments, by directly using pre-trained models on Cifar10, CelebA and LSUN, PNDMs can generate higher quality synthetic images with only 50 steps compared with 1000-step DDIMs (20x speedup), significantly outperform DDIMs with 250 steps (by around 0.4 in FID) and have good generalization on different variance schedules. Our implementation is available at https://github.com/luping-liu/PNDM. <|reference_end|>", "<|reference_start|> Progressive Distillation for Fast Sampling of Diffusion Models: Diffusion models have recently shown great promise for generative modeling, outperforming GANs on perceptual quality and autoregressive models at density estimation. A remaining downside is their slow sampling time: generating high quality samples takes many hundreds or thousands of model evaluations. Here we make two contributions to help eliminate this downside: First, we present new parameterizations of diffusion models that provide increased stability when using few sampling steps. Second, we present a method to distill a trained deterministic diffusion sampler, using many steps, into a new diffusion model that takes half as many sampling steps. We then keep progressively applying this distillation procedure to our model, halving the number of required sampling steps each time. On standard image generation benchmarks like CIFAR-10, ImageNet, and LSUN, we start out with state-of-the-art samplers taking as many as 8192 steps, and are able to distill down to models taking as few as 4 steps without losing much perceptual quality; achieving, for example, a FID of 3.0 on CIFAR-10 in 4 steps. Finally, we show that the full progressive distillation procedure does not take more time than it takes to train the original model, thus representing an efficient solution for generative modeling using diffusion at both train and test time. <|reference_end|>", "<|reference_start|> Auto-Encoding Variational Bayes: How can we perform efficient inference and learning in directed probabilistic models, in the presence of continuous latent variables with intractable posterior distributions, and large datasets? We introduce a stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiability conditions, even works in the intractable case. Our contributions are two-fold. First, we show that a reparameterization of the variational lower bound yields a lower bound estimator that can be straightforwardly optimized using standard stochastic gradient methods. Second, we show that for i.i.d. datasets with continuous latent variables per datapoint, posterior inference can be made especially efficient by fitting an approximate inference model (also called a recognition model) to the intractable posterior using the proposed lower bound estimator. Theoretical advantages are reflected in experimental results. <|reference_end|>", "<|reference_start|> Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference: Latent Diffusion models (LDMs) have achieved remarkable results in synthesizing high-resolution images. However, the iterative sampling process is computationally intensive and leads to slow generation. Inspired by Consistency Models (song et al.), we propose Latent Consistency Models (LCMs), enabling swift inference with minimal steps on any pre-trained LDMs, including Stable Diffusion (rombach et al). Viewing the guided reverse diffusion process as solving an augmented probability flow ODE (PF-ODE), LCMs are designed to directly predict the solution of such ODE in latent space, mitigating the need for numerous iterations and allowing rapid, high-fidelity sampling. Efficiently distilled from pre-trained classifier-free guided diffusion models, a high-quality 768 x 768 2~4-step LCM takes only 32 A100 GPU hours for training. Furthermore, we introduce Latent Consistency Fine-tuning (LCF), a novel method that is tailored for fine-tuning LCMs on customized image datasets. Evaluation on the LAION-5B-Aesthetics dataset demonstrates that LCMs achieve state-of-the-art text-to-image generation performance with few-step inference. Project Page: https://latent-consistency-models.github.io/ <|reference_end|>" ]

[ 21, 23, 29, 41 ]

[ "<|reference_start|> Gemini: A Family of Highly Capable Multimodal Models: This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI. <|reference_end|>", "<|reference_start|> Learning Transferable Visual Models From Natural Language Supervision: State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories. This restricted form of supervision limits their generality and usability since additional labeled data is needed to specify any other visual concept. Learning directly from raw text about images is a promising alternative which leverages a much broader source of supervision. We demonstrate that the simple pre-training task of predicting which caption goes with which image is an efficient and scalable way to learn SOTA image representations from scratch on a dataset of 400 million (image, text) pairs collected from the internet. After pre-training, natural language is used to reference learned visual concepts (or describe new ones) enabling zero-shot transfer of the model to downstream tasks. We study the performance of this approach by benchmarking on over 30 different existing computer vision datasets, spanning tasks such as OCR, action recognition in videos, geo-localization, and many types of fine-grained object classification. The model transfers non-trivially to most tasks and is often competitive with a fully supervised baseline without the need for any dataset specific training. For instance, we match the accuracy of the original ResNet-50 on ImageNet zero-shot without needing to use any of the 1.28 million training examples it was trained on. We release our code and pre-trained model weights at https://github.com/OpenAI/CLIP. <|reference_end|>", "<|reference_start|> GPT-4 Technical Report: We report the development of GPT-4, a large-scale, multimodal model which can accept image and text inputs and produce text outputs. While less capable than humans in many real-world scenarios, GPT-4 exhibits human-level performance on various professional and academic benchmarks, including passing a simulated bar exam with a score around the top 10% of test takers. GPT-4 is a Transformer-based model pre-trained to predict the next token in a document. The post-training alignment process results in improved performance on measures of factuality and adherence to desired behavior. A core component of this project was developing infrastructure and optimization methods that behave predictably across a wide range of scales. This allowed us to accurately predict some aspects of GPT-4's performance based on models trained with no more than 1/1,000th the compute of GPT-4. <|reference_end|>", "<|reference_start|> InFoBench: Evaluating Instruction Following Ability in Large Language Models: This paper introduces the Decomposed Requirements Following Ratio (DRFR), a new metric for evaluating Large Language Models' (LLMs) ability to follow instructions. Addressing a gap in current methodologies, DRFR breaks down complex instructions into simpler criteria, facilitating a detailed analysis of LLMs' compliance with various aspects of tasks. Alongside this metric, we present InFoBench, a benchmark comprising 500 diverse instructions and 2,250 decomposed questions across multiple constraint categories. Our experiments compare DRFR with traditional scoring methods and explore annotation sources, including human experts, crowd-sourced workers, and GPT-4. The findings demonstrate DRFR's higher reliability and the effectiveness of using GPT-4 as a cost-efficient annotator. The evaluation of several advanced LLMs using this framework reveals their strengths and areas needing improvement, particularly in complex instruction-following. This study contributes a novel metric and benchmark, offering insights for future LLM development and evaluation. <|reference_end|>" ]

[ 3, 6, 8, 26 ]

[ "<|reference_start|> Distributed Space-Time-Coded Protocols for Exploiting Cooperative Diversity in Wireless Networks: We develop and analyze space-time coded cooperative diversity protocols for combating multipath fading across multiple protocol layers in a wireless network. The protocols exploit spatial diversity available among a collection of distributed terminals that relay messages for one another in such a manner that the destination terminal can average the fading, even though it is unknown a priori which terminals will be involved. In particular, a source initiates transmission to its destination, and many relays potentially receive the transmission. Those terminals that can fully decode the transmission utilize a space-time code to cooperatively relay to the destination. We demonstrate that these protocols achieve full spatial diversity in the number of cooperating terminals, not just the number of decoding relays, and can be used effectively for higher spectral efficiencies than repetition-based schemes. We discuss issues related to space-time code design for these protocols, emphasizing codes that readily allow for appealing distributed versions. <|reference_end|>", "<|reference_start|> A Simple Cooperative Diversity Method Based on Network Path Selection: Cooperative diversity has been recently proposed as a way to form virtual antenna arrays that provide dramatic gains in slow fading wireless environments. However most of the proposed solutions require distributed space-time coding algorithms, the careful design of which is left for future investigation if there is more than one cooperative relay. We propose a novel scheme, that alleviates these problems and provides diversity gains on the order of the number of relays in the network. Our scheme first selects the best relay from a set of M available relays and then uses this best relay for cooperation between the source and the destination. We develop and analyze a distributed method to select the best relay that requires no topology information and is based on local measurements of the instantaneous channel conditions. This method also requires no explicit communication among the relays. The success (or failure) to select the best available path depends on the statistics of the wireless channel, and a methodology to evaluate performance for any kind of wireless channel statistics, is provided. Information theoretic analysis of outage probability shows that our scheme achieves the same diversity-multiplexing tradeoff as achieved by more complex protocols, where coordination and distributed space-time coding for M nodes is required, such as those proposed in [7]. The simplicity of the technique, allows for immediate implementation in existing radio hardware and its adoption could provide for improved flexibility, reliability and efficiency in future 4G wireless systems. <|reference_end|>", "<|reference_start|> Full length article: Wireless transmission using cooperation on demand: <|reference_end|>", "<|reference_start|> Performance analysis of single relay selection in Rayleigh fading: We provide closed-form expressions for the outage and bit error probability (BEP) of uncoded, threshold-based opportunistic relaying (OR) and selection cooperation (SC), at arbitrary signal to noise ratios (SNRs) and number of available relays, assuming decode-and-forward relays and Rayleigh fading channels. Numerical results demonstrate that SC performs slightly better in terms of outage probability; in terms of BEP, both systems may outperform one another, depending on the SNR threshold that determines the set of relays that participate in the forwarding process. <|reference_end|>" ]

[ 0, 2, 3, 15 ]

[ "<|reference_start|> Recurrent computations for visual pattern completion: Making inferences from partial information constitutes a critical aspect of cognition. During visual perception, pattern completion enables recognition of poorly visible or occluded objects. We combined psychophysics, physiology and computational models to test the hypothesis that pattern completion is implemented by recurrent computations and present three pieces of evidence that are consistent with this hypothesis. First, subjects robustly recognized objects even when rendered <15% visible, but recognition was largely impaired when processing was interrupted by backward masking. Second, invasive physiological responses along the human ventral cortex exhibited visually selective responses to partially visible objects that were delayed compared to whole objects, suggesting the need for additional computations. These physiological delays were correlated with the effects of backward masking. Third, state-of-the-art feed-forward computational architectures were not robust to partial visibility. However, recognition performance was recovered when the model was augmented with attractor-based recurrent connectivity. These results provide a strong argument of plausibility for the role of recurrent computations in making visual inferences from partial information. <|reference_end|>", "<|reference_start|> Recurrent computations for visual pattern completion: Making inferences from partial information constitutes a critical aspect of cognition. During visual perception, pattern completion enables recognition of poorly visible or occluded objects. We combined psychophysics, physiology and computational models to test the hypothesis that pattern completion is implemented by recurrent computations and present three pieces of evidence that are consistent with this hypothesis. First, subjects robustly recognized objects even when rendered <15% visible, but recognition was largely impaired when processing was interrupted by backward masking. Second, invasive physiological responses along the human ventral cortex exhibited visually selective responses to partially visible objects that were delayed compared to whole objects, suggesting the need for additional computations. These physiological delays were correlated with the effects of backward masking. Third, state-of-the-art feed-forward computational architectures were not robust to partial visibility. However, recognition performance was recovered when the model was augmented with attractor-based recurrent connectivity. These results provide a strong argument of plausibility for the role of recurrent computations in making visual inferences from partial information. <|reference_end|>", "<|reference_start|> Neurons with graded response have collective computational properties like those of two-state neurons: A model for a large network of \"neurons\" with a graded response (or sigmoid input-output relation) is studied. This deterministic system has collective properties in very close correspondence with the earlier stochastic model based on McCulloch - Pitts neurons. The content- addressable memory and other emergent collective properties of the original model also are present in the graded response model. The idea that such collective properties are used in biological systems is given added credence by the continued presence of such properties for more nearly biological \"neurons.\" Collective analog electrical circuits of the kind described will certainly function. The collective states of the two models have a simple correspondence. The original model will continue to be useful for simulations, because its connection to graded response systems is established. Equations that include the effect of action potentials in the graded response system are also developed. <|reference_end|>", "<|reference_start|> Dense Associative Memory for Pattern Recognition: A model of associative memory is studied, which stores and reliably retrieves many more patterns than the number of neurons in the network. We propose a simple duality between this dense associative memory and neural networks commonly used in deep learning. On the associative memory side of this duality, a family of models that smoothly interpolates between two limiting cases can be constructed. One limit is referred to as the feature-matching mode of pattern recognition, and the other one as the prototype regime. On the deep learning side of the duality, this family corresponds to feedforward neural networks with one hidden layer and various activation functions, which transmit the activities of the visible neurons to the hidden layer. This family of activation functions includes logistics, rectified linear units, and rectified polynomials of higher degrees. The proposed duality makes it possible to apply energy-based intuition from associative memory to analyze computational properties of neural networks with unusual activation functions - the higher rectified polynomials which until now have not been used in deep learning. The utility of the dense memories is illustrated for two test cases: the logical gate XOR and the recognition of handwritten digits from the MNIST data set. <|reference_end|>" ]

[ 0, 1, 4, 5 ]

[ "<|reference_start|> Deep Unsupervised Learning using Nonequilibrium Thermodynamics: A central problem in machine learning involves modeling complex data-sets using highly flexible families of probability distributions in which learning, sampling, inference, and evaluation are still analytically or computationally tractable. Here, we develop an approach that simultaneously achieves both flexibility and tractability. The essential idea, inspired by non-equilibrium statistical physics, is to systematically and slowly destroy structure in a data distribution through an iterative forward diffusion process. We then learn a reverse diffusion process that restores structure in data, yielding a highly flexible and tractable generative model of the data. This approach allows us to rapidly learn, sample from, and evaluate probabilities in deep generative models with thousands of layers or time steps, as well as to compute conditional and posterior probabilities under the learned model. We additionally release an open source reference implementation of the algorithm. <|reference_end|>", "<|reference_start|> Attention U-Net: Learning Where to Look for the Pancreas: We propose a novel attention gate (AG) model for medical imaging that automatically learns to focus on target structures of varying shapes and sizes. Models trained with AGs implicitly learn to suppress irrelevant regions in an input image while highlighting salient features useful for a specific task. This enables us to eliminate the necessity of using explicit external tissue/organ localisation modules of cascaded convolutional neural networks (CNNs). AGs can be easily integrated into standard CNN architectures such as the U-Net model with minimal computational overhead while increasing the model sensitivity and prediction accuracy. The proposed Attention U-Net architecture is evaluated on two large CT abdominal datasets for multi-class image segmentation. Experimental results show that AGs consistently improve the prediction performance of U-Net across different datasets and training sizes while preserving computational efficiency. The code for the proposed architecture is publicly available. <|reference_end|>", "<|reference_start|> High-Resolution Image Synthesis with Latent Diffusion Models: By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond. Additionally, their formulation allows for a guiding mechanism to control the image generation process without retraining. However, since these models typically operate directly in pixel space, optimization of powerful DMs often consumes hundreds of GPU days and inference is expensive due to sequential evaluations. To enable DM training on limited computational resources while retaining their quality and flexibility, we apply them in the latent space of powerful pretrained autoencoders. In contrast to previous work, training diffusion models on such a representation allows for the first time to reach a near-optimal point between complexity reduction and detail preservation, greatly boosting visual fidelity. By introducing cross-attention layers into the model architecture, we turn diffusion models into powerful and flexible generators for general conditioning inputs such as text or bounding boxes and high-resolution synthesis becomes possible in a convolutional manner. Our latent diffusion models (LDMs) achieve a new state of the art for image inpainting and highly competitive performance on various tasks, including unconditional image generation, semantic scene synthesis, and super-resolution, while significantly reducing computational requirements compared to pixel-based DMs. Code is available at https://github.com/CompVis/latent-diffusion . <|reference_end|>", "<|reference_start|> Generating Diverse High-Fidelity Images with VQ-VAE-2: We explore the use of Vector Quantized Variational AutoEncoder (VQ-VAE) models for large scale image generation. To this end, we scale and enhance the autoregressive priors used in VQ-VAE to generate synthetic samples of much higher coherence and fidelity than possible before. We use simple feed-forward encoder and decoder networks, making our model an attractive candidate for applications where the encoding and/or decoding speed is critical. Additionally, VQ-VAE requires sampling an autoregressive model only in the compressed latent space, which is an order of magnitude faster than sampling in the pixel space, especially for large images. We demonstrate that a multi-scale hierarchical organization of VQ-VAE, augmented with powerful priors over the latent codes, is able to generate samples with quality that rivals that of state of the art Generative Adversarial Networks on multifaceted datasets such as ImageNet, while not suffering from GAN's known shortcomings such as mode collapse and lack of diversity. <|reference_end|>" ]

[ 14, 15, 17, 38 ]

[ "<|reference_start|> End-to-{{End: Driven by the aggressive scaling of modern IC technologies, network-on-chip (NoC) becomes increasingly susceptible to various noise sources. In this paper, adaptive error correction code injection scheme is represented to achieve both high reliability and low latency in various temperature condition. Simulation results show that the latency of proposed scheme is 50∼57% of the 2G4L code without(with smaller) reliability degradation. <|reference_end|>", "<|reference_start|> PALR: Personalization Aware LLMs for Recommendation: Large language models (LLMs) have recently received significant attention for their exceptional capabilities. Despite extensive efforts in developing general-purpose LLMs that can be utilized in various natural language processing (NLP) tasks, there has been less research exploring their potential in recommender systems. In this paper, we propose a novel framework, named PALR, which aiming to combine user history behaviors (such as clicks, purchases, ratings, etc.) with LLMs to generate user preferred items. Specifically, we first use user/item interactions as guidance for candidate retrieval. Then we adopt a LLM-based ranking model to generate recommended items. Unlike existing approaches that typically adopt general-purpose LLMs for zero/few-shot recommendation testing or training on small-sized language models (with less than 1 billion parameters), which cannot fully elicit LLMs' reasoning abilities and leverage rich item side parametric knowledge, we fine-tune a 7 billion parameters LLM for the ranking purpose. This model takes retrieval candidates in natural language format as input, with instruction which explicitly asking to select results from input candidates during inference. Our experimental results demonstrate that our solution outperforms state-of-the-art models on various sequential recommendation tasks. <|reference_end|>", "<|reference_start|> Rethinking Personalized Ranking at Pinterest: An End-to-End Approach: In this work, we present our journey to revolutionize the personalized recommendation engine through end-to-end learning from raw user actions. We encode user's long-term interest in Pinner- Former, a user embedding optimized for long-term future actions via a new dense all-action loss, and capture user's short-term intention by directly learning from the real-time action sequences. We conducted both offline and online experiments to validate the performance of the new model architecture, and also address the challenge of serving such a complex model using mixed CPU/GPU setup in production. The proposed system has been deployed in production at Pinterest and has delivered significant online gains across organic and Ads applications. <|reference_end|>", "<|reference_start|> Personalized Response Generation via Generative Split Memory Network: Despite the impressive successes of generation and dialogue systems, how to endow a text generation system with particular personality traits to deliver more personalized responses remains under-investigated. In this work, we look at how to generate personalized responses for questions on Reddit by utilizing personalized user profiles and posting histories. Specifically, we release an open-domain single-turn dialog dataset made up of 1.5M conversation pairs together with 300k profiles of users and related comments. We then propose a memory network to generate personalized responses in dialogue that utilizes a novel mechanism of splitting memories: one for user profile meta attributes and the other for user-generated information like comment histories. Experimental results show the quantitative and qualitative improvements of our simple split memory network model over the state-of-the-art response generation baselines. <|reference_end|>" ]

[ 4, 8, 9, 15 ]

[ "<|reference_start|> {Harvesting Robots for High-value Crops: State-of-the-art Review and Challenges Ahead: This review article analyzes state‐of‐the‐art and future perspectives for harvesting robots in high‐value crops. The objectives were to characterize the crop environment relevant for robotic harvesting, to perform a literature review on the state‐of‐the‐art of harvesting robots using quantitative measures, and to reflect on the crop environment and literature review to formulate challenges and directions for future research and development. Harvesting robots were reviewed regarding the crop harvested in a production environment, performance indicators, design process techniques used, hardware design decisions, and algorithm characteristics. On average, localization success was 85%, detachment success was 75%, harvest success was 66%, fruit damage was 5%, peduncle damage was 45%, and cycle time was 33 s. A kiwi harvesting robot achieved the shortest cycle time of 1 s. Moreover, the performance of harvesting robots did not improve in the past three decades, and none of these 50 robots was commercialized. Four future challenges with R&D directions were identified to realize a positive trend in performance and to successfully implement harvesting robots in practice: (1) simplifying the task, (2) enhancing the robot, (3) defining requirements and measuring performance, and (4) considering additional requirements for successful implementation. This review article may provide new directions for future automation projects in high‐value crops. <|reference_end|>", "<|reference_start|> Selective Harvesting Robotics: Current Research, Trends, and Future Directions: <|reference_end|>", "<|reference_start|> {Performance evaluation of a harvesting robot for sweet pepper: This paper evaluates a robot developed for autonomous harvesting of sweet peppers in a commercial greenhouse. Objectives were to assess robot performance under unmodified and simplified crop conditions, using two types of end effectors (Fin Ray; Lip type), and to evaluate the performance contribution of stem‐dependent determination of the grasp pose. We describe and discuss the performance of hardware and software components developed for fruit harvesting in a complex environment that includes lighting variation, occlusions, and densely spaced obstacles. After simplifying the crop, harvest success significantly improved from 6% to 26% (Fin Ray) and from 2% to 33% (Lip type). We observed a decrease in stem damage and an increase in grasp success after enabling stem‐dependent determination of the grasp pose. Generally, the robot had difficulty in successfully picking sweet peppers and we discuss possible causes. The robot's novel capability of perceiving the stem of a plant may serve as useful functionality for future robots. <|reference_end|>", "<|reference_start|> Mixtures of Lightweight Deep Convolutional Neural Networks: Applied to Agricultural Robotics: We propose a novel approach for training deep convolutional neural networks (DCNNs) that allows us to tradeoff complexity and accuracy to learn lightweight models suitable for robotic platforms such as AgBot II (which performs automated weed management). Our approach consists of three stages, the first is to adapt a pre-trained model to the task at hand. This provides state-of-the-art performance but at the cost of high computational complexity resulting in a low frame rate of just 0.12 frames per second (fps). Second, we use the adapted model and employ model compression techniques to learn a lightweight DCNN that is less accurate but has two orders of magnitude fewer parameters. Third, $K$ lightweight models are combined as a mixture model to further enhance the performance of the lightweight models. Applied to the challenging task of weed segmentation, we improve the accuracy from 85.9%, using a traditional approach, to 93.9% by adapting a complicated pre-trained DCNN with 25M parameters (Inception-v3). The downside to this adapted model, Adapted-IV3, is that it can only process 0.12 fps. To make this approach fast while still retaining accuracy, we learn lightweight DCNNs which when combined can achieve accuracy greater than 90% while using considerably fewer parameters capable of processing between 1.07 and 1.83 fps, up to an order of magnitude faster and up to an order of magnitude fewer parameters. <|reference_end|>" ]

[ 9, 10, 17, 21 ]

[ "<|reference_start|> False Positives, False Negatives, and False Analyses: A Rejoinder to \"Machine Bias: There's Software Used across the Country to Predict Future Criminals. and It's Biased against Blacks\": PROPUBLICA RECENTLY RELEASED a much-heralded investigative report claim­ ing that a risk assessment tool (known as the COMPAS) used in criminal justice is biased against black defendants.12 The report heavily implied that such bias is inherent in all actuarial risk assessment instruments (ARAIs). We think ProPublica’s report was based on faulty statistics and data analysis, and that the report failed to show that the COMPAS itself is racially biased, let alone that other risk instruments are biased. Not only do ProPublica’s results contradict several com­ prehensive existing studies concluding that actuarial risk can be predicted free of racial <|reference_end|>", "<|reference_start|> False Positives, False Negatives, and False Analyses: A Rejoinder to \"Machine Bias: There's Software Used across the Country to Predict Future Criminals. and It's Biased against Blacks\": PROPUBLICA RECENTLY RELEASED a much-heralded investigative report claim­ ing that a risk assessment tool (known as the COMPAS) used in criminal justice is biased against black defendants.12 The report heavily implied that such bias is inherent in all actuarial risk assessment instruments (ARAIs). We think ProPublica’s report was based on faulty statistics and data analysis, and that the report failed to show that the COMPAS itself is racially biased, let alone that other risk instruments are biased. Not only do ProPublica’s results contradict several com­ prehensive existing studies concluding that actuarial risk can be predicted free of racial <|reference_end|>", "<|reference_start|> {Actionable Auditing: Investigating the Impact of Publicly Naming Biased Performance Results of Commercial AI Products: Although algorithmic auditing has emerged as a key strategy to expose systematic biases embedded in software platforms, we struggle to understand the real-world impact of these audits, as scholarship on the impact of algorithmic audits on increasing algorithmic fairness and transparency in commercial systems is nascent. To analyze the impact of publicly naming and disclosing performance results of biased AI systems, we investigate the commercial impact of Gender Shades, the first algorithmic audit of gender and skin type performance disparities in commercial facial analysis models. This paper 1) outlines the audit design and structured disclosure procedure used in the Gender Shades study, 2) presents new performance metrics from targeted companies IBM, Microsoft and Megvii (Face++) on the Pilot Parliaments Benchmark (PPB) as of August 2018, 3) provides performance results on PPB by non-target companies Amazon and Kairos and, 4) explores differences in company responses as shared through corporate communications that contextualize differences in performance on PPB. Within 7 months of the original audit, we find that all three targets released new API versions. All targets reduced accuracy disparities between males and females and darker and lighter-skinned subgroups, with the most significant update occurring for the darker-skinned female subgroup, that underwent a 17.7% - 30.4% reduction in error between audit periods. Minimizing these disparities led to a 5.72% to 8.3% reduction in overall error on the Pilot Parliaments Benchmark (PPB) for target corporation APIs. The overall performance of non-targets Amazon and Kairos lags significantly behind that of the targets, with error rates of 8.66% and 6.60% overall, and error rates of 31.37% and 22.50% for the darker female subgroup, respectively. <|reference_end|>", "<|reference_start|> Discrimination in the Age of Algorithms: The law forbids discrimination. But the ambiguity of human decision-making often makes it extraordinarily hard for the legal system to know whether anyone has actually discriminated. To understand how algorithms affect discrimination, we must therefore also understand how they affect the problem of detecting discrimination. By one measure, algorithms are fundamentally opaque, not just cognitively but even mathematically. Yet for the task of proving discrimination, processes involving algorithms can provide crucial forms of transparency that are otherwise unavailable. These benefits do not happen automatically. But with appropriate requirements in place, the use of algorithms will make it possible to more easily examine and interrogate the entire decision process, thereby making it far easier to know whether discrimination has occurred. By forcing a new level of specificity, the use of algorithms also highlights, and makes transparent, central tradeoffs among competing values. Algorithms are not only a threat to be regulated; with the right safeguards in place, they have the potential to be a positive force for equity. <|reference_end|>" ]

[ 1, 2, 3, 7 ]

[ "<|reference_start|> Automated cardiovascular magnetic resonance image analysis with fully convolutional networks: Cardiovascular magnetic resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR image analysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images. Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV) end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV). By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance on par with human experts in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. <|reference_end|>", "<|reference_start|> Few-Shot Adversarial Learning of Realistic Neural Talking Head Models: Several recent works have shown how highly realistic human head images can be obtained by training convolutional neural networks to generate them. In order to create a personalized talking head model, these works require training on a large dataset of images of a single person. However, in many practical scenarios, such personalized talking head models need to be learned from a few image views of a person, potentially even a single image. Here, we present a system with such few-shot capability. It performs lengthy meta-learning on a large dataset of videos, and after that is able to frame few- and one-shot learning of neural talking head models of previously unseen people as adversarial training problems with high capacity generators and discriminators. Crucially, the system is able to initialize the parameters of both the generator and the discriminator in a person-specific way, so that training can be based on just a few images and done quickly, despite the need to tune tens of millions of parameters. We show that such an approach is able to learn highly realistic and personalized talking head models of new people and even portrait paintings. <|reference_end|>", "<|reference_start|> AdaptIS: Adaptive Instance Selection Network: We present Adaptive Instance Selection network architecture for class-agnostic instance segmentation. Given an input image and a point $(x, y)$, it generates a mask for the object located at $(x, y)$. The network adapts to the input point with a help of AdaIN layers, thus producing different masks for different objects on the same image. AdaptIS generates pixel-accurate object masks, therefore it accurately segments objects of complex shape or severely occluded ones. AdaptIS can be easily combined with standard semantic segmentation pipeline to perform panoptic segmentation. To illustrate the idea, we perform experiments on a challenging toy problem with difficult occlusions. Then we extensively evaluate the method on panoptic segmentation benchmarks. We obtain state-of-the-art results on Cityscapes and Mapillary even without pretraining on COCO, and show competitive results on a challenging COCO dataset. The source code of the method and the trained models are available at https://github.com/saic-vul/adaptis. <|reference_end|>", "<|reference_start|> FiLM: Visual Reasoning with a General Conditioning Layer: We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning - answering image-related questions which require a multi-step, high-level process - a task which has proven difficult for standard deep learning methods that do not explicitly model reasoning. Specifically, we show on visual reasoning tasks that FiLM layers 1) halve state-of-the-art error for the CLEVR benchmark, 2) modulate features in a coherent manner, 3) are robust to ablations and architectural modifications, and 4) generalize well to challenging, new data from few examples or even zero-shot. <|reference_end|>" ]

[ 0, 9, 10, 11 ]

[ "<|reference_start|> Undecidability on quantum finite automata: Our model in this paper is a 1.5.way quantum finite automaton which can move its head 0 or +1 position but not -1 position. It is shown that the most fundamental decision question, the emptiness problem, is not solvable for this model. Note that the emptiness problem is solvable far push-down automata and even for one-way nondeterministic stack automata. <|reference_end|>", "<|reference_start|> A time complexity gap for two-way probabilistic finite-state automata: It is shown that if a two-way probabilistic finite-state automaton (2pfa) M recognizes a nonregular language L with error probability bounded below $\\frac{1}{2}$, then there is a positive constant b (depending on M) such that, for infinitely many inputs x, the expected running time of M on input x must exceed $2^{n^{b}}$ where n is the length of x. This complements a result of Freivalds showing that 2pfa’s can recognize certain nonregular languages in exponential expected time. It also establishes a time complexity gap for 2pfa’s, since any regular language can be recognized by some 2pfa in linear time. Other results give roughly exponential upper and lower bounds on the worst-case increase in the number of states when converting a polynomial-time 2pfa to an equivalent two-way nondeterministic finite-state automaton or to an equivalent one-way deterministic finite-state automaton. <|reference_end|>", "<|reference_start|> One-way probabilistic reversible and quantum one-counter automata: <|reference_end|>", "<|reference_start|> Unbounded-error quantum computation with small space bounds: We prove the following facts about the language recognition power of quantum Turing machines (QTMs) in the unbounded error setting: QTMs are strictly more powerful than probabilistic Turing machines for any common space bound $ s $ satisfying $ s(n)=o(\\log \\log n) $. For \"one-way\" Turing machines, where the input tape head is not allowed to move left, the above result holds for $s(n)=o(\\log n) $. We also give a characterization for the class of languages recognized with unbounded error by real-time quantum finite automata (QFAs) with restricted measurements. It turns out that these automata are equal in power to their probabilistic counterparts, and this fact does not change when the QFA model is augmented to allow general measurements and mixed states. Unlike the case with classical finite automata, when the QFA tape head is allowed to remain stationary in some steps, more languages become recognizable. We define and use a QTM model that generalizes the other variants introduced earlier in the study of quantum space complexity. <|reference_end|>" ]

[ 0, 1, 11, 15 ]

[ "<|reference_start|> Traffic modelling of asynchronous bufferless optical packet switched networks: <|reference_end|>", "<|reference_start|> Traffic models for slotted optical packet switched networks: <|reference_end|>", "<|reference_start|> Network coding, Fundamentals and Applications: Thank you very much for downloading network coding fundamentals and applications. As you may know, people have search hundreds times for their chosen readings like this network coding fundamentals and applications, but end up in malicious downloads. Rather than enjoying a good book with a cup of tea in the afternoon, instead they cope with some malicious bugs inside their laptop. network coding fundamentals and applications is available in our book collection an online access to it is set as public so you can download it instantly. Our digital library hosts in multiple countries, allowing you to get the most less latency time to download any of our books like this one. Merely said, the network coding fundamentals and applications is universally compatible with any devices to read. <|reference_end|>", "<|reference_start|> Combined study on survivability and performance in optical packet switched networks: Survivability and performance constitute crucial quality of service (QoS) issues in future optical packet switched (OPS) networks. I take an integrated view on survivability and performance in OPS networks by presenting the extended shared packet redundancy scheme (ESPRS). The ESPRS combines shared packet redundancy with 1+1 path protection functionality. I focus on the performance of the ESPRS in different failure situations and show how the packet loss rate is influenced by the number of node- and link-disjoint paths between a node pair, the loss probability on the paths, number of failures, the relative amount of redundancy, and the size of the packet set. An analytical model of the ESPRS is provided. <|reference_end|>" ]

[ 3, 4, 8, 9 ]

[ "<|reference_start|> Multimodal Machine Learning: A Survey and Taxonomy: Our experience of the world is multimodal - we see objects, hear sounds, feel texture, smell odors, and taste flavors. Modality refers to the way in which something happens or is experienced and a research problem is characterized as multimodal when it includes multiple such modalities. In order for Artificial Intelligence to make progress in understanding the world around us, it needs to be able to interpret such multimodal signals together. Multimodal machine learning aims to build models that can process and relate information from multiple modalities. It is a vibrant multi-disciplinary field of increasing importance and with extraordinary potential. Instead of focusing on specific multimodal applications, this paper surveys the recent advances in multimodal machine learning itself and presents them in a common taxonomy. We go beyond the typical early and late fusion categorization and identify broader challenges that are faced by multimodal machine learning, namely: representation, translation, alignment, fusion, and co-learning. This new taxonomy will enable researchers to better understand the state of the field and identify directions for future research. <|reference_end|>", "<|reference_start|> A Comprehensive Survey on Cross-modal Retrieval: In recent years, cross-modal retrieval has drawn much attention due to the rapid growth of multimodal data. It takes one type of data as the query to retrieve relevant data of another type. For example, a user can use a text to retrieve relevant pictures or videos. Since the query and its retrieved results can be of different modalities, how to measure the content similarity between different modalities of data remains a challenge. Various methods have been proposed to deal with such a problem. In this paper, we first review a number of representative methods for cross-modal retrieval and classify them into two main groups: 1) real-valued representation learning, and 2) binary representation learning. Real-valued representation learning methods aim to learn real-valued common representations for different modalities of data. To speed up the cross-modal retrieval, a number of binary representation learning methods are proposed to map different modalities of data into a common Hamming space. Then, we introduce several multimodal datasets in the community, and show the experimental results on two commonly used multimodal datasets. The comparison reveals the characteristic of different kinds of cross-modal retrieval methods, which is expected to benefit both practical applications and future research. Finally, we discuss open problems and future research directions. <|reference_end|>", "<|reference_start|> CRET: Cross-Modal Retrieval Transformer for Efficient Text-Video Retrieval: Given a text query, the text-to-video retrieval task aims to find the relevant videos in the database. Recently, model-based (MDB) methods have demonstrated superior accuracy than embedding-based (EDB) methods due to their excellent capacity of modeling local video/text correspondences, especially when equipped with large-scale pre-training schemes like ClipBERT. Generally speaking, MDB methods take a text-video pair as input and harness deep models to predict the mutual similarity, while EDB methods first utilize modality-specific encoders to extract embeddings for text and video, then evaluate the distance based on the extracted embeddings. Notably, MDB methods cannot produce explicit representations for text and video, instead, they have to exhaustively pair the query with every database item to predict their mutual similarities in the inference stage, which results in significant inefficiency in practical applications. In this work, we propose a novel EDB method CRET (Cross-modal REtrieval Transformer), which not only demonstrates promising efficiency in retrieval tasks, but also achieves better accuracy than existing MDB methods. The credits are mainly attributed to our proposed Cross-modal Correspondence Modeling (CCM) module and Gaussian Estimation of Embedding Space (GEES) loss. Specifically, the CCM module is composed by transformer decoders and a set of decoder centers. With the help of the learned decoder centers, the text/video embeddings can be efficiently aligned, without suffering from pairwise model-based inference. Moreover, to balance the information loss and computational overhead when sampling frames from a given video, we present a novel GEES loss, which implicitly conducts dense sampling in the video embedding space, without suffering from heavy computational cost. Extensive experiments show that without pre-training on extra datasets, our proposed CRET outperforms the state-of-the-art MDB methods that were pre-trained on additional datasets, meanwhile still shows promising efficiency in retrieval tasks. <|reference_end|>", "<|reference_start|> Cross-Modal Learning: Adaptivity, Prediction and Interaction: <|reference_end|>" ]

[ 0, 3, 4, 6 ]

[ "<|reference_start|> Big Self-Supervised Models Advance Medical Image Classification: Self-supervised pretraining followed by supervised fine-tuning has seen success in image recognition, especially when labeled examples are scarce, but has received limited attention in medical image analysis. This paper studies the effectiveness of self-supervised learning as a pretraining strategy for medical image classification. We conduct experiments on two distinct tasks: dermatology skin condition classification from digital camera images and multi-label chest X-ray classification, and demonstrate that self-supervised learning on ImageNet, followed by additional self-supervised learning on unlabeled domain-specific medical images significantly improves the accuracy of medical image classifiers. We introduce a novel Multi-Instance Contrastive Learning (MICLe) method that uses multiple images of the underlying pathology per patient case, when available, to construct more informative positive pairs for self-supervised learning. Combining our contributions, we achieve an improvement of 6.7% in top-1 accuracy and an improvement of 1.1% in mean AUC on dermatology and chest X-ray classification respectively, outperforming strong supervised baselines pretrained on ImageNet. In addition, we show that big self-supervised models are robust to distribution shift and can learn efficiently with a small number of labeled medical images. <|reference_end|>", "<|reference_start|> Transformers in Medical Imaging: A Survey: Following unprecedented success on the natural language tasks, Transformers have been successfully applied to several computer vision problems, achieving state-of-the-art results and prompting researchers to reconsider the supremacy of convolutional neural networks (CNNs) as {de facto} operators. Capitalizing on these advances in computer vision, the medical imaging field has also witnessed growing interest for Transformers that can capture global context compared to CNNs with local receptive fields. Inspired from this transition, in this survey, we attempt to provide a comprehensive review of the applications of Transformers in medical imaging covering various aspects, ranging from recently proposed architectural designs to unsolved issues. Specifically, we survey the use of Transformers in medical image segmentation, detection, classification, reconstruction, synthesis, registration, clinical report generation, and other tasks. In particular, for each of these applications, we develop taxonomy, identify application-specific challenges as well as provide insights to solve them, and highlight recent trends. Further, we provide a critical discussion of the field's current state as a whole, including the identification of key challenges, open problems, and outlining promising future directions. We hope this survey will ignite further interest in the community and provide researchers with an up-to-date reference regarding applications of Transformer models in medical imaging. Finally, to cope with the rapid development in this field, we intend to regularly update the relevant latest papers and their open-source implementations at \\url{https://github.com/fahadshamshad/awesome-transformers-in-medical-imaging}. <|reference_end|>", "<|reference_start|> Interpretable deep learning systems for multi-class segmentation and classification of non-melanoma skin cancer: <|reference_end|>", "<|reference_start|> Concept Bottleneck Models: We seek to learn models that we can interact with using high-level concepts: if the model did not think there was a bone spur in the x-ray, would it still predict severe arthritis? State-of-the-art models today do not typically support the manipulation of concepts like \"the existence of bone spurs\", as they are trained end-to-end to go directly from raw input (e.g., pixels) to output (e.g., arthritis severity). We revisit the classic idea of first predicting concepts that are provided at training time, and then using these concepts to predict the label. By construction, we can intervene on these concept bottleneck models by editing their predicted concept values and propagating these changes to the final prediction. On x-ray grading and bird identification, concept bottleneck models achieve competitive accuracy with standard end-to-end models, while enabling interpretation in terms of high-level clinical concepts (\"bone spurs\") or bird attributes (\"wing color\"). These models also allow for richer human-model interaction: accuracy improves significantly if we can correct model mistakes on concepts at test time. <|reference_end|>" ]

[ 1, 2, 11, 13 ]

[ "<|reference_start|> Treatment and dosimetric advantages between VMAT, IMRT, and helical tomotherapy in prostate cancer.: <|reference_end|>", "<|reference_start|> On Convergence and Stability of GANs: We propose studying GAN training dynamics as regret minimization, which is in contrast to the popular view that there is consistent minimization of a divergence between real and generated distributions. We analyze the convergence of GAN training from this new point of view to understand why mode collapse happens. We hypothesize the existence of undesirable local equilibria in this non-convex game to be responsible for mode collapse. We observe that these local equilibria often exhibit sharp gradients of the discriminator function around some real data points. We demonstrate that these degenerate local equilibria can be avoided with a gradient penalty scheme called DRAGAN. We show that DRAGAN enables faster training, achieves improved stability with fewer mode collapses, and leads to generator networks with better modeling performance across a variety of architectures and objective functions. <|reference_end|>", "<|reference_start|> Automatic chemical design using a data-driven continuous representation of molecules: We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous representations also allow the use of powerful gradient-based optimization to efficiently guide the search for optimized functional compounds. We demonstrate our method in the domain of drug-like molecules and also in the set of molecules with fewer that nine heavy atoms. <|reference_end|>", "<|reference_start|> High-dimensional Bayesian optimization using low-dimensional feature spaces: Bayesian optimization (BO) is a powerful approach for seeking the global optimum of expensive black-box functions and has proven successful for fine tuning hyper-parameters of machine learning models. However, BO is practically limited to optimizing 10--20 parameters. To scale BO to high dimensions, we usually make structural assumptions on the decomposition of the objective and\\slash or exploit the intrinsic lower dimensionality of the problem, e.g. by using linear projections. We could achieve a higher compression rate with nonlinear projections, but learning these nonlinear embeddings typically requires much data. This contradicts the BO objective of a relatively small evaluation budget. To address this challenge, we propose to learn a low-dimensional feature space jointly with (a) the response surface and (b) a reconstruction mapping. Our approach allows for optimization of BO's acquisition function in the lower-dimensional subspace, which significantly simplifies the optimization problem. We reconstruct the original parameter space from the lower-dimensional subspace for evaluating the black-box function. For meaningful exploration, we solve a constrained optimization problem. <|reference_end|>" ]

[ 6, 22, 25, 29 ]

[ "<|reference_start|> Extracting Information about Security Vulnerabilities from Web Text: The Web is an important source of information about computer security threats, vulnerabilities and cyber attacks. We present initial work on developing a framework to detect and extract information about vulnerabilities and attacks from Web text. Our prototype system uses Wikitology, a general purpose knowledge base derived from Wikipedia, to extract concepts that describe specific vulnerabilities and attacks, map them to related concepts from DBpedia and generate machine understandable assertions. Such a framework will be useful in adding structure to already existing vulnerability descriptions as well as detecting new ones. We evaluate our approach against vulnerability descriptions from the National Vulnerability Database. Our results suggest that it can be useful in monitoring streams of text from social media or chat rooms to identify potential new attacks and vulnerabilities or to collect data on the spread and volume of existing ones. <|reference_end|>", "<|reference_start|> Toward never ending language learning: We report research toward a never-ending language learning system, focusing on a first implementation which learns to classify occurrences of noun phrases according to lexical categories such as “city” and “university.” Our experiments suggest that the accuracy of classifiers produced by semi-supervised learning can be improved by coupling the learning of multiple classes based on background knowledge about relationships between the classes (e.g., ”university” is mutually exclusive of ”company”, and is a subset of ”organization”). <|reference_end|>", "<|reference_start|> Extracting Patterns and Relations from the World Wide Web: <|reference_end|>", "<|reference_start|> Coupled semi-supervised learning for information extraction: We consider the problem of semi-supervised learning to extract categories (e.g., academic fields, athletes) and relations (e.g., PlaysSport(athlete, sport)) from web pages, starting with a handful of labeled training examples of each category or relation, plus hundreds of millions of unlabeled web documents. Semi-supervised training using only a few labeled examples is typically unreliable because the learning task is underconstrained. This paper pursues the thesis that much greater accuracy can be achieved by further constraining the learning task, by coupling the semi-supervised training of many extractors for different categories and relations. We characterize several ways in which the training of category and relation extractors can be coupled, and present experimental results demonstrating significantly improved accuracy as a result. <|reference_end|>" ]

[ 1, 10, 11, 17 ]

[ "<|reference_start|> Stretch-shortening cycle: a powerful model to study normal and fatigued muscle.: <|reference_end|>", "<|reference_start|> {Individualized Muscle-Tendon Assessment and Training: The interaction of muscle and tendon is of major importance for movement performance and a balanced development of muscle strength and tendon stiffness could protect athletes from overuse injury. However, muscle and tendon do not necessarily adapt in a uniform manner during a training process. The development of a diagnostic routine to assess both the strength capacity of muscle and the mechanical properties of tendons would enable the detection of muscle-tendon imbalances, indicate if the training should target muscle strength or tendon stiffness development and allow for the precise prescription of training loads to optimize tendon adaptation. This perspective article discusses a framework of individualized muscle-tendon assessment and training and outlines a methodological approach for the patellar tendon. <|reference_end|>", "<|reference_start|> {Medial gastrocnemius and soleus muscle-tendon unit, fascicle, and tendon interaction during walking in children with cerebral palsy: This study investigates the in vivo function of the medial gastrocnemius and soleus muscle‐tendon units (MTU), fascicles, and tendons during walking in children with cerebral palsy (CP) and an equinus gait pattern. <|reference_end|>", "<|reference_start|> Optical flow estimation using high frame rate sequences: Gradient-based optical flow estimation methods such as the Lucas-Kanade (1981) method work well for scenes with small displacements but fail when objects move with large displacements. Hierarchical matching-based methods do not suffer from large displacements but are less accurate. By utilizing the high speed imaging capability of CMOS image sensors, the frame rate can be increased to obtain more accurate optical flow with wide range of scene velocities in real time. Further, by integrating the memory and processing with the sensor on the same chip, optical flow estimation using high frame rate sequences can be performed without unduly increasing the off-chip data rate. The paper describes a method for obtaining high accuracy optical flow at a standard frame rate using high frame rate sequences. The Lucas-Kanade method is used to obtain optical flow estimates at high frame rate, which are then accumulated and refined to obtain optical flow estimates at a standard frame rate. The method is tested on video sequences synthetically generated by perspective warping. The results demonstrate significant improvements in optical flow estimation accuracy with moderate memory and computational power requirements. <|reference_end|>" ]

[ 0, 5, 15, 23 ]

[ "<|reference_start|> A medical mirror for non-contact health monitoring: Digital medical devices promise to transform the future of medicine because of their ability to produce exquisitely detailed individual physiological data. As ordinary people start to have access and control over their own physiological data, they can play a more active role in the management of their health. This revolution must take place in our everyday lives, not just in the doctor's office or research lab. However, current techniques for physiological monitoring typically require users to strap on bulky sensors, chest straps or sticky electrodes. This discourages regular use because the sensors can be uncomfortable or encumbering. In this work, we propose a new mirror interface for real-time, contact-free measurements of heart rate without the need for external sensors. Users can have the experience of remote health monitoring by simply looking into the Medical Mirror. <|reference_end|>", "<|reference_start|> Robust pulse rate from chrominance-based rPPG: Remote photoplethysmography (rPPG) enables contactless monitoring of the blood volume pulse using a regular camera. Recent research focused on improved motion robustness, but the proposed blind source separation techniques (BSS) in RGB color space show limited success. We present an analysis of the motion problem, from which far superior chrominance-based methods emerge. For a population of 117 stationary subjects, we show our methods to perform in 92% good agreement (±1.96σ) with contact PPG, with RMSE and standard deviation both a factor of 2 better than BSS-based methods. In a fitness setting using a simple spectral peak detector, the obtained pulse-rate for modest motion (bike) improves from 79% to 98% correct, and for vigorous motion (stepping) from less than 11% to more than 48% correct. We expect the greatly improved robustness to considerably widen the application scope of the technology. <|reference_end|>", "<|reference_start|> Robust pulse rate from chrominance-based rPPG: Remote photoplethysmography (rPPG) enables contactless monitoring of the blood volume pulse using a regular camera. Recent research focused on improved motion robustness, but the proposed blind source separation techniques (BSS) in RGB color space show limited success. We present an analysis of the motion problem, from which far superior chrominance-based methods emerge. For a population of 117 stationary subjects, we show our methods to perform in 92% good agreement (±1.96σ) with contact PPG, with RMSE and standard deviation both a factor of 2 better than BSS-based methods. In a fitness setting using a simple spectral peak detector, the obtained pulse-rate for modest motion (bike) improves from 79% to 98% correct, and for vigorous motion (stepping) from less than 11% to more than 48% correct. We expect the greatly improved robustness to considerably widen the application scope of the technology. <|reference_end|>", "<|reference_start|> Heart rate estimation from facial videos using a spatiotemporal representation with convolutional neural networks: Remote photoplethysmography (rPPG) is a kind of noncontact technique to measure heart rate (HR) from facial videos. As the demand for long-term health monitoring grows, rPPG attracts much attention from researchers. However, the performance of conventional rPPG methods is easily degenerated due to noise interference. Recently, some deep learning-based rPPG methods have been introduced and they revealed good performance against noise. In this article, we propose a new rPPG method with convolutional neural networks (CNNs) to build a mapping between a spatiotemporal HR feature image to its corresponding HR value. The feature map is constructed in a time-delayed way with noise-contaminated pulse signals extracted from existing rPPG methods. The CNN model is trained using transfer learning where images built from synthetic rPPG signals are taken to train the model first in order to generate initials for the practical one. The synthetic rPPG signals are interpolated from blood volume pulses or electrocardiograms through a modified Akima cubic Hermite interpolation. The proposed method is tested in both within-database and cross-database configurations on public databases. The results demonstrate that our method achieves overall the best performance compared to some other typical rPPG methods. The mean absolute error reaches 5.98 beats per minute and the mean error rate percentage is 7.97% in the cross-database testing on MAHNOB-HCI data set. Besides, some key factors that affect the performance of our method are also discussed which indicates potential ways for further improvements. <|reference_end|>" ]

[ 9, 15, 21, 30 ]

[ "<|reference_start|> {Unsupervised Learning of Narrative Schemas and Their Participants: We describe an unsupervised system for learning narrative schemas, coherent sequences or sets of events (arrested(POLICE, SUSPECT), convicted(JUDGE, SUSPECT)) whose arguments are filled with participant semantic roles defined over words (Judge = {judge, jury, court}, Police = {police, agent, authorities}). Unlike most previous work in event structure or semantic role learning, our system does not use supervised techniques, hand-built knowledge, or predefined classes of events or roles. Our unsupervised learning algorithm uses coreferring arguments in chains of verbs to learn both rich narrative event structure and argument roles. By jointly addressing both tasks, we improve on previous results in narrative/frame learning and induce rich frame-specific semantic roles. <|reference_end|>", "<|reference_start|> Knowledgeable reader: Enhancing cloze-style reading comprehension with external commonsense knowledge.: We introduce a neural reading comprehension model that integrates external commonsense knowledge, encoded as a key-value memory, in a cloze-style setting. Instead of relying only on document-to-question interaction or discrete features as in prior work, our model attends to relevant external knowledge and combines this knowledge with the context representation before inferring the answer. This allows the model to attract and imply knowledge from an external knowledge source that is not explicitly stated in the text, but that is relevant for inferring the answer. Our model improves results over a very strong baseline on a hard Common Nouns dataset, making it a strong competitor of much more complex models. By including knowledge explicitly, our model can also provide evidence about the background knowledge used in the RC process. <|reference_end|>", "<|reference_start|> Story cloze task: Uw nlp system: This paper describes University of Washington NLP’s submission for the Linking Models of Lexical, Sentential and Discourse-level Semantics (LSDSem 2017) shared task—the Story Cloze Task. Our system is a linear classifier with a variety of features, including both the scores of a neural language model and style features. We report 75.2% accuracy on the task. A further discussion of our results can be found in Schwartz et al. (2017). <|reference_end|>", "<|reference_start|> Story Comprehension for Predicting What Happens Next: Automatic story comprehension is a fundamental challenge in Natural Language Understanding, and can enable computers to learn about social norms, human behavior and commonsense. In this paper, we present a story comprehension model that explores three distinct semantic aspects: (i) the sequence of events described in the story, (ii) its emotional trajectory, and (iii) its plot consistency. We judge the model’s understanding of real-world stories by inquiring if, like humans, it can develop an expectation of what will happen next in a given story. Specifically, we use it to predict the correct ending of a given short story from possible alternatives. The model uses a hidden variable to weigh the semantic aspects in the context of the story. Our experiments demonstrate the potential of our approach to characterize these semantic aspects, and the strength of the hidden variable based approach. The model outperforms the state-of-the-art approaches and achieves best results on a publicly available dataset. <|reference_end|>" ]

[ 2, 35, 36, 37 ]

[ "<|reference_start|> {The Link-prediction Problem for Social Networks: Given a snapshot of a social network, can we infer which new interactions among its members are likely to occur in the near future? We formalize this question as the link prediction problem, and develop approaches to link prediction based on measures the \"proximity\" of nodes in a network. Experiments on large co-authorship networks suggest that information about future interactions can be extracted from network topology alone, and that fairly subtle measures for detecting node proximity can outperform more direct measures. <|reference_end|>", "<|reference_start|> Network completion and survey sampling: We study the problem of learning the topology of an undirected network by observing a random subsample. Specifically, the sample is chosen by randomly selecting a fixed number of vertices, and for each we are allowed to observe all edges it is incident with. We analyze a general formalization of learning from such samples, and derive confidence bounds on the number of differences between the true and learned topologies, as a function of the number of observed mistakes and the algorithm’s bias. In addition to this general analysis, we also analyze a variant of the problem under a stochastic block model assumption. <|reference_end|>", "<|reference_start|> Inferring Networks of Diffusion and Influence: Information diffusion and virus propagation are fundamental processes taking place in networks. While it is often possible to directly observe when nodes become infected with a virus or adopt the information, observing individual transmissions (i.e., who infects whom, or who influences whom) is typically very difficult. Furthermore, in many applications, the underlying network over which the diffusions and propagations spread is actually unobserved. We tackle these challenges by developing a method for tracing paths of diffusion and influence through networks and inferring the networks over which contagions propagate. Given the times when nodes adopt pieces of information or become infected, we identify the optimal network that best explains the observed infection times. Since the optimization problem is NP-hard to solve exactly, we develop an efficient approximation algorithm that scales to large datasets and finds provably near-optimal networks. We demonstrate the effectiveness of our approach by tracing information diffusion in a set of 170 million blogs and news articles over a one year period to infer how information flows through the online media space. We find that the diffusion network of news for the top 1,000 media sites and blogs tends to have a core-periphery structure with a small set of core media sites that diffuse information to the rest of the Web. These sites tend to have stable circles of influence with more general news media sites acting as connectors between them. <|reference_end|>", "<|reference_start|> Inferring Networks of Diffusion and Influence: Information diffusion and virus propagation are fundamental processes taking place in networks. While it is often possible to directly observe when nodes become infected with a virus or adopt the information, observing individual transmissions (i.e., who infects whom, or who influences whom) is typically very difficult. Furthermore, in many applications, the underlying network over which the diffusions and propagations spread is actually unobserved. We tackle these challenges by developing a method for tracing paths of diffusion and influence through networks and inferring the networks over which contagions propagate. Given the times when nodes adopt pieces of information or become infected, we identify the optimal network that best explains the observed infection times. Since the optimization problem is NP-hard to solve exactly, we develop an efficient approximation algorithm that scales to large datasets and finds provably near-optimal networks. We demonstrate the effectiveness of our approach by tracing information diffusion in a set of 170 million blogs and news articles over a one year period to infer how information flows through the online media space. We find that the diffusion network of news for the top 1,000 media sites and blogs tends to have a core-periphery structure with a small set of core media sites that diffuse information to the rest of the Web. These sites tend to have stable circles of influence with more general news media sites acting as connectors between them. <|reference_end|>" ]

[ 4, 8, 17, 18 ]

[ "<|reference_start|> Topology discovery for network fault management using mobile agents in ad-hoc networks: Managing today's complex and increasingly heterogeneous networks requires in-depth knowledge and extensive training as well as collection of very large amount of data. Fault management is one of the functional areas of network management that entails detection, identification and correction of anomalies that disrupt services of a network. The task of fault management is even harder in ad-hoc networks where the topology of the network changes frequently. It is very inefficient if not impossible to discover the ad-hoc network topology using traditional practices of network discovery. We propose a mobile multi agent system for topology discovery that will allow fault management functions in ad-hoc network. Comparison to current mobile agent based topology discovery systems is also presented <|reference_end|>", "<|reference_start|> Fast Detection of Compact Topology Representation for Wireless Networks: This paper considers a hybrid cellular architecture in which mobiles can communicate with others in their vicinity, e.g. using 802.11 interface, in addition to the base stations of the cellular network. Such an architecture can aid device-to-device communication as well as assist critical tasks of cellular networks such as mobility management, content caching and relaying. In order to enable these capabilities, base stations need to have sufficient knowledge of the underlying network topology induced by the 802.11 links of the mobiles. Due to the dynamic nature of this network, a compressed snapshot of its topology should be collected within a very short time duration and with minimal communication among mobiles. Addressing this need, we propose a compact topology representation that is suitable for a number of applications. We utilize the broadcast nature of wireless channels to design an efficient topology detection algorithm that acquires a compact representation of the underlying network (at most 3N links and a low `stretch' factor for the N mobiles) within a short duration (10s of ms). Our scheme does not have collision resolution, backoffs or any of the other MAC layer inefficiencies. <|reference_end|>", "<|reference_start|> Design and Evaluation of a Hybrid D2D Discovery Mechanism in 5G Cellular Networks: Device-to-device (D2D) communications allow devices to communicate directly without using base stations for relaying, yielding several merits such as increasing the spectral efficiency and system capacity, reducing network latency and power consumption of UEs, offloading the 5G cellular networks, and extending the network coverage. However, to realize D2D communications, the first challenging issue is how to efficiently find target devices in proximity for communications. Therefore, in this paper we intend to design a hybrid D2D discovery mechanism and evaluate its performance. Simulation results demonstrate that the proposed D2D discovery mechanism achieves better performance in terms of power consumption and discovery time, compared with the direct discovery scheme that purely utilizes the contention-based transmission scheme and operates in the unlicensed band. <|reference_end|>", "<|reference_start|> Learning the Interference Graph of a Wireless Network: A key challenge in wireless networking is the management of interference between transmissions. Identifying which transmitters interfere with each other is a crucial first step. In this paper we cast the task of estimating the a wireless interference environment as a graph learning problem. Nodes represent transmitters and edges represent the presence of interference between pairs of transmitters. We passively observe network traffic transmission patterns and collect information on transmission successes and failures. We establish bounds on the number of observations (each a snapshot of a network traffic pattern) required to identify the interference graph reliably with high probability. Our main results are scaling laws that tell us how the number of observations must grow in terms of the total number of nodes $n$ in the network and the maximum number of interfering transmitters $d$ per node (maximum node degree). The effects of hidden terminal interference (i.e., interference not detectable via carrier sensing) on the observation requirements are also quantified. We show that to identify the graph it is necessary and sufficient that the observation period grows like $d^2 \\log n$, and we propose a practical algorithm that reliably identifies the graph from this length of observation. The observation requirements scale quite mildly with network size, and networks with sparse interference (small $d$) can be identified more rapidly. Computational experiments based on a realistic simulations of the traffic and protocol lend additional support to these conclusions. <|reference_end|>" ]

[ 1, 2, 3, 4 ]

[ "<|reference_start|> Hegel’s {Dialectics: “Dialectics” is a term used to describe a method of philosophical argument that involves some sort of contradictory process between opposing sides. In what is perhaps the most classic version of “dialectics”, the ancient Greek philosopher, Plato (see entry on Plato), for instance, presented his philosophical argument as a back-and-forth dialogue or debate, generally between the character of Socrates, on one side, and some person or group of people to whom Socrates was talking (his interlocutors), on the other. In the course of the dialogues, Socrates’ interlocutors propose definitions of philosophical concepts or express views that Socrates challenges or opposes. The back-and-forth debate between opposing sides produces a kind of linear progression or evolution in philosophical views or positions: as the dialogues go along, Socrates’ interlocutors change or refine their views in response to Socrates’ challenges and come to adopt more sophisticated views. The back-and-forth dialectic between Socrates and his interlocutors thus becomes Plato’s way of arguing against the earlier, less sophisticated views or positions and for the more sophisticated ones later. <|reference_end|>", "<|reference_start|> Cortex: A Compiler for Recursive Deep Learning Models: Optimizing deep learning models is generally performed in two steps: (i) high-level graph optimizations such as kernel fusion and (ii) low level kernel optimizations such as those found in vendor libraries. This approach often leaves significant performance on the table, especially for the case of recursive deep learning models. In this paper, we present Cortex, a compiler-based approach to generate highly-efficient code for recursive models for low latency inference. Our compiler approach and low reliance on vendor libraries enables us to perform end-to-end optimizations, leading to up to 14X lower inference latencies over past work, across different backends. <|reference_end|>", "<|reference_start|> Explaining and Harnessing Adversarial Examples: Several machine learning models, including neural networks, consistently misclassify adversarial examples---inputs formed by applying small but intentionally worst-case perturbations to examples from the dataset, such that the perturbed input results in the model outputting an incorrect answer with high confidence. Early attempts at explaining this phenomenon focused on nonlinearity and overfitting. We argue instead that the primary cause of neural networks' vulnerability to adversarial perturbation is their linear nature. This explanation is supported by new quantitative results while giving the first explanation of the most intriguing fact about them: their generalization across architectures and training sets. Moreover, this view yields a simple and fast method of generating adversarial examples. Using this approach to provide examples for adversarial training, we reduce the test set error of a maxout network on the MNIST dataset. <|reference_end|>", "<|reference_start|> Hyper-parameterized Dialectic Search for Non-linear Box-Constrained Optimization with Heterogenous Variable Types: <|reference_end|>" ]

[ 2, 5, 6, 10 ]

[ "<|reference_start|> {Search-Based Program Synthesis: A promising, useful tool for future programming development environments. <|reference_end|>", "<|reference_start|> Synthesis of Data Completion Scripts using Finite Tree Automata: In application domains that store data in a tabular format, a common task is to fill the values of some cells using values stored in other cells. For instance, such data completion tasks arise in the context of missing value imputation in data science and derived data computation in spreadsheets and relational databases. Unfortunately, end-users and data scientists typically struggle with many data completion tasks that require non-trivial programming expertise. This paper presents a synthesis technique for automating data completion tasks using programming-by-example (PBE) and a very lightweight sketching approach. Given a formula sketch (e.g., AVG($?_1$, $?_2$)) and a few input-output examples for each hole, our technique synthesizes a program to automate the desired data completion task. Towards this goal, we propose a domain-specific language (DSL) that combines spatial and relational reasoning over tabular data and a novel synthesis algorithm that can generate DSL programs that are consistent with the input-output examples. The key technical novelty of our approach is a new version space learning algorithm that is based on finite tree automata (FTA). The use of FTAs in the learning algorithm leads to a more compact representation that allows more sharing between programs that are consistent with the examples. We have implemented the proposed approach in a tool called DACE and evaluate it on 84 benchmarks taken from online help forums. We also illustrate the advantages of our approach by comparing our technique against two existing synthesizers, namely PROSE and SKETCH. <|reference_end|>", "<|reference_start|> FlashRelate: extracting relational data from semi-structured spreadsheets using examples: With hundreds of millions of users, spreadsheets are one of the most important end-user applications. Spreadsheets are easy to use and allow users great flexibility in storing data. This flexibility comes at a price: users often treat spreadsheets as a poor man's database, leading to creative solutions for storing high-dimensional data. The trouble arises when users need to answer queries with their data. Data manipulation tools make strong assumptions about data layouts and cannot read these ad-hoc databases. Converting data into the appropriate layout requires programming skills or a major investment in manual reformatting. The effect is that a vast amount of real-world data is \"locked-in\" to a proliferation of one-off formats. We introduce FlashRelate, a synthesis engine that lets ordinary users extract structured relational data from spreadsheets without programming. Instead, users extract data by supplying examples of output relational tuples. FlashRelate uses these examples to synthesize a program in Flare. Flare is a novel extraction language that extends regular expressions with geometric constructs. An interactive user interface on top of FlashRelate lets end users extract data by point-and-click. We demonstrate that correct Flare programs can be synthesized in seconds from a small set of examples for 43 real-world scenarios. Finally, our case study demonstrates FlashRelate's usefulness addressing the widespread problem of data trapped in corporate and government formats. <|reference_end|>", "<|reference_start|> Synthesizing Bijective Lenses: Bidirectional transformations between different data representations occur frequently in modern software systems. They appear as serializers and deserializers, as database views and view updaters, and more. Manually building bidirectional transformations---by writing two separate functions that are intended to be inverses---is tedious and error prone. A better approach is to use a domain-specific language in which both directions can be written as a single expression. However, these domain-specific languages can be difficult to program in, requiring programmers to manage fiddly details while working in a complex type system. To solve this, we present Optician, a tool for type-directed synthesis of bijective string transformers. The inputs to Optician are two ordinary regular expressions representing two data formats and a few concrete examples for disambiguation. The output is a well-typed program in Boomerang (a bidirectional language based on the theory of lenses). The main technical challenge involves navigating the vast program search space efficiently enough. Unlike most prior work on type-directed synthesis, our system operates in the context of a language with a rich equivalence relation on types (the theory of regular expressions). We synthesize terms of a equivalent language and convert those generated terms into our lens language. We prove the correctness of our synthesis algorithm. We also demonstrate empirically that our new language changes the synthesis problem from one that admits intractable solutions to one that admits highly efficient solutions. We evaluate Optician on a benchmark suite of 39 examples including both microbenchmarks and realistic examples derived from other data management systems including Flash Fill, a tool for synthesizing string transformations in spreadsheets, and Augeas, a tool for bidirectional processing of Linux system configuration files. <|reference_end|>" ]

[ 3, 4, 5, 7 ]

[ "<|reference_start|> In Advances in Neural Information Processing Systems: <|reference_end|>", "<|reference_start|> In Advances in Neural Information Processing Systems: <|reference_end|>", "<|reference_start|> Algorithms For Reinforcement Learning Synthesis Lectures On Artificial Intelligence And Machine Learning: The articles presented here were selected from preliminary versions presented at the International Conference on Genetic Algorithms in June 1991, as well as at a special Workshop on Genetic Algorithms for Machine Learning at the same Conference. Genetic algorithms are general-purpose search algorithms that use principles inspired by natural population genetics to evolve solutions to problems. The basic idea is to maintain a population of knowledge structure that represent candidate solutions to the problem of interest. The population evolves over time through a process of competition (i.e. survival of the fittest) and controlled variation (i.e. recombination and mutation). Genetic Algorithms for Machine Learning contains articles on three topics that have not been the focus of many previous articles on GAs, namely concept learning from examples, reinforcement learning for control, and theoretical analysis of GAs. It is hoped that this sample will serve to broaden the acquaintance of the general machine learning community with the major areas of work on GAs. The articles in this book address a number of central issues in applying GAs to machine learning problems. For example, the choice of appropriate representation and the corresponding set of genetic learning operators is an important set of decisions facing a user of a genetic algorithm. The study of genetic algorithms is proceeding at a robust pace. If experimental progress and theoretical understanding continue to evolve as expected, genetic algorithms will continue to provide a distinctive approach to machine learning. Genetic Algorithms for Machine Learning is an edited volume of original research made up of invited contributions by leading researchers. This book is Open Access under a CC BY licence. The LNCS 11427 and 11428 proceedings set constitutes the proceedings of the 25th International Conference on Tools and Algorithms for the Construction and Analysis of Systems, TACAS 2019, which took place in Prague, Czech Republic, in April 2019, held as part of the European Joint Conferences on Theory and Practice of Software, ETAPS 2019. The total of 42 full and 8 short tool demo papers presented in these volumes was carefully reviewed and selected from 164 submissions. The papers are organized in topical sections as follows: Part I: SAT and SMT, SAT solving and theorem proving; verification and analysis; model checking; tool demo; and machine learning. Part II: concurrent and distributed systems; monitoring and runtime verification; hybrid and stochastic <|reference_end|>", "<|reference_start|> DEUP: Direct Epistemic Uncertainty Prediction: Epistemic Uncertainty is a measure of the lack of knowledge of a learner which diminishes with more evidence. While existing work focuses on using the variance of the Bayesian posterior due to parameter uncertainty as a measure of epistemic uncertainty, we argue that this does not capture the part of lack of knowledge induced by model misspecification. We discuss how the excess risk, which is the gap between the generalization error of a predictor and the Bayes predictor, is a sound measure of epistemic uncertainty which captures the effect of model misspecification. We thus propose a principled framework for directly estimating the excess risk by learning a secondary predictor for the generalization error and subtracting an estimate of aleatoric uncertainty, i.e., intrinsic unpredictability. We discuss the merits of this novel measure of epistemic uncertainty, and highlight how it differs from variance-based measures of epistemic uncertainty and addresses its major pitfall. Our framework, Direct Epistemic Uncertainty Prediction (DEUP) is particularly interesting in interactive learning environments, where the learner is allowed to acquire novel examples in each round. Through a wide set of experiments, we illustrate how existing methods in sequential model optimization can be improved with epistemic uncertainty estimates from DEUP, and how DEUP can be used to drive exploration in reinforcement learning. We also evaluate the quality of uncertainty estimates from DEUP for probabilistic image classification and predicting synergies of drug combinations. <|reference_end|>" ]

[ 6, 17, 25, 39 ]

[ "<|reference_start|> SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size: Recent research on deep neural networks has focused primarily on improving accuracy. For a given accuracy level, it is typically possible to identify multiple DNN architectures that achieve that accuracy level. With equivalent accuracy, smaller DNN architectures offer at least three advantages: (1) Smaller DNNs require less communication across servers during distributed training. (2) Smaller DNNs require less bandwidth to export a new model from the cloud to an autonomous car. (3) Smaller DNNs are more feasible to deploy on FPGAs and other hardware with limited memory. To provide all of these advantages, we propose a small DNN architecture called SqueezeNet. SqueezeNet achieves AlexNet-level accuracy on ImageNet with 50x fewer parameters. Additionally, with model compression techniques we are able to compress SqueezeNet to less than 0.5MB (510x smaller than AlexNet). \nThe SqueezeNet architecture is available for download here: this https URL <|reference_end|>", "<|reference_start|> A Simple Weight Decay Can Improve Generalization: It has been observed in numerical simulations that a weight decay can improve generalization in a feed-forward neural network. This paper explains why. It is proven that a weight decay has two effects in a linear network. First, it suppresses any irrelevant components of the weight vector by choosing the smallest vector that solves the learning problem. Second, if the size is chosen right, a weight decay can suppress some of the effects of static noise on the targets, which improves generalization quite a lot. It is then shown how to extend these results to networks with hidden layers and non-linear units. Finally the theory is confirmed by some numerical simulations using the data from NetTalk. <|reference_end|>", "<|reference_start|> Convolutional Pose Machines: Pose Machines provide a sequential prediction framework for learning rich implicit spatial models. In this work we show a systematic design for how convolutional networks can be incorporated into the pose machine framework for learning image features and image-dependent spatial models for the task of pose estimation. The contribution of this paper is to implicitly model long-range dependencies between variables in structured prediction tasks such as articulated pose estimation. We achieve this by designing a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference. Our approach addresses the characteristic difficulty of vanishing gradients during training by providing a natural learning objective function that enforces intermediate supervision, thereby replenishing back-propagated gradients and conditioning the learning procedure. We demonstrate state-of-the-art performance and outperform competing methods on standard benchmarks including the MPII, LSP, and FLIC datasets. <|reference_end|>", "<|reference_start|> Multiscale Deep Equilibrium Models: We propose a new class of implicit networks, the multiscale deep equilibrium model (MDEQ), suited to large-scale and highly hierarchical pattern recognition domains. An MDEQ directly solves for and backpropagates through the equilibrium points of multiple feature resolutions simultaneously, using implicit differentiation to avoid storing intermediate states (and thus requiring only $O(1)$ memory consumption). These simultaneously-learned multi-resolution features allow us to train a single model on a diverse set of tasks and loss functions, such as using a single MDEQ to perform both image classification and semantic segmentation. We illustrate the effectiveness of this approach on two large-scale vision tasks: ImageNet classification and semantic segmentation on high-resolution images from the Cityscapes dataset. In both settings, MDEQs are able to match or exceed the performance of recent competitive computer vision models: the first time such performance and scale have been achieved by an implicit deep learning approach. The code and pre-trained models are at https://github.com/locuslab/mdeq . <|reference_end|>" ]

[ 7, 35, 57, 64 ]

[ "<|reference_start|> Mastering the game of Go with deep neural networks and tree search: <|reference_end|>", "<|reference_start|> Dota 2 with Large Scale Deep Reinforcement Learning: On April 13th, 2019, OpenAI Five became the first AI system to defeat the world champions at an esports game. The game of Dota 2 presents novel challenges for AI systems such as long time horizons, imperfect information, and complex, continuous state-action spaces, all challenges which will become increasingly central to more capable AI systems. OpenAI Five leveraged existing reinforcement learning techniques, scaled to learn from batches of approximately 2 million frames every 2 seconds. We developed a distributed training system and tools for continual training which allowed us to train OpenAI Five for 10 months. By defeating the Dota 2 world champion (Team OG), OpenAI Five demonstrates that self-play reinforcement learning can achieve superhuman performance on a difficult task. <|reference_end|>", "<|reference_start|> AW-Opt: Learning Robotic Skills with Imitation and Reinforcement at Scale: Robotic skills can be learned via imitation learning (IL) using user-provided demonstrations, or via reinforcement learning (RL) using large amountsof autonomously collected experience.Both methods have complementarystrengths and weaknesses: RL can reach a high level of performance, but requiresexploration, which can be very time consuming and unsafe; IL does not requireexploration, but only learns skills that are as good as the provided demonstrations.Can a single method combine the strengths of both approaches? A number ofprior methods have aimed to address this question, proposing a variety of tech-niques that integrate elements of IL and RL. However, scaling up such methodsto complex robotic skills that integrate diverse offline data and generalize mean-ingfully to real-world scenarios still presents a major challenge. In this paper, ouraim is to test the scalability of prior IL + RL algorithms and devise a system basedon detailed empirical experimentation that combines existing components in themost effective and scalable way. To that end, we present a series of experimentsaimed at understanding the implications of each design decision, so as to develop acombined approach that can utilize demonstrations and heterogeneous prior datato attain the best performance on a range of real-world and realistic simulatedrobotic problems. Our complete method, which we call AW-Opt, combines ele-ments of advantage-weighted regression [1, 2] and QT-Opt [3], providing a unifiedapproach for integrating demonstrations and offline data for robotic manipulation.Please see https://awopt.github.io for more details. <|reference_end|>", "<|reference_start|> IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures: In this work we aim to solve a large collection of tasks using a single reinforcement learning agent with a single set of parameters. A key challenge is to handle the increased amount of data and extended training time. We have developed a new distributed agent IMPALA (Importance Weighted Actor-Learner Architecture) that not only uses resources more efficiently in single-machine training but also scales to thousands of machines without sacrificing data efficiency or resource utilisation. We achieve stable learning at high throughput by combining decoupled acting and learning with a novel off-policy correction method called V-trace. We demonstrate the effectiveness of IMPALA for multi-task reinforcement learning on DMLab-30 (a set of 30 tasks from the DeepMind Lab environment (Beattie et al., 2016)) and Atari-57 (all available Atari games in Arcade Learning Environment (Bellemare et al., 2013a)). Our results show that IMPALA is able to achieve better performance than previous agents with less data, and crucially exhibits positive transfer between tasks as a result of its multi-task approach. <|reference_end|>" ]

[ 0, 2, 4, 10 ]

[ "<|reference_start|> 2014 47th Annual IEEE/ACM International Symposium on Microarchitecture: <|reference_end|>", "<|reference_start|> 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA): <|reference_end|>", "<|reference_start|> 51st Annual IEEE/ACM International Symposium on Microarchitecture, MICRO 2018, Fukuoka, Japan, October 20-24, 2018: <|reference_end|>", "<|reference_start|> International Conference on Learning Representations (ICLR): <|reference_end|>" ]

[ 4, 13, 17, 53 ]

[ "<|reference_start|> Traffic engineering with MPLS: From the Publisher: \nOptimize network bandwidth with Traffic Engineering and MPLS \n \nHard to find information on how to use MPLS traffic engineering to optimize network bandwidth, save on network cost, and improve customer satisfactionUnderstand the theoretical underpinnings of the various protocols that comprise traffic engineeringLearn basic and advanced configuration of traffic engineering and related services like QoS and ATM interaction Suggested network designs, configuration examples, and an end-to-end case study provide readers with practical, working solutions readers can implement on their own networks \n \nAs corporations seek to reduce costs, improve efficiencies, gain market share and profit they increasingly are looking to their own information technology systems as a means to this end. Traffic engineering allows engineers to maximize network resources. It resolves the issue of having large amount of traffic on certain portions of the network while other portions go under-utilized. Traffic Engineering with MPLS provides readers with information on how to use MPLS traffic engineering and associated features to optimize network bandwidth. The book covers forwarding fundamentals, traffic engineering theory, protocol descriptions, deployment guidelines, configuration, show commands, and debugs. This book is a one-stop reference for understanding MPLS traffic engineering and implementing it on the network. A comprehensive case study is used to show a complete MPLS traffic engineering deployment. <|reference_end|>", "<|reference_start|> Load balancing for multiple traffic matrices using sdn hybrid routing: Classical traffic engineering (TE) methods calculate the optimal routing based on a single traffic matrix. However, they are unable to handle unexpected traffic changes. Thus, it is of interest to find a good routing configuration to accommodate multiple possible traffic scenarios. There are two major approaches to achieve load balancing for multiple traffic matrices: destination-based routing and explicit routing. It has been shown that explicit routing performs better than destination-based routing for multiple traffic matrices. However, explicit routing has high complexity and requires large Ternary Content Addressable Memory (TCAM) in the routers. Thus, it is power hungry and unscalable. This paper presents an approach called hybrid routing to achieve load balancing for multiple traffic matrices with low complexity and good scalability. Our basic idea is to complement destination-based routing with a small number of explicit routing forwarding entries to take advantage of both two routing approaches. Hybrid routing greatly reduces the number of forwarding entries compared with pure explicit routing. This has great value for practice in that the scheme requires very small TCAM to implement. Hybrid routing is very suitable for implementation using SDN. A heuristic algorithm is developed to obtain the near-optimal hybrid routing configuration. Extensive evaluation demonstrates the effectiveness of hybrid routing. The results show that hybrid routing achieves near-optimal load balancing compared with pure explicit routing. In particular, hybrid routing saves at least 84.6% TCAM resources in all practical networks used in our evaluation. <|reference_end|>", "<|reference_start|> OSPF Optimized Multipath (OSPF-OMP): <|reference_end|>", "<|reference_start|> Experience-driven Networking: A Deep Reinforcement Learning based Approach: Modern communication networks have become very complicated and highly dynamic, which makes them hard to model, predict and control. In this paper, we develop a novel experience-driven approach that can learn to well control a communication network from its own experience rather than an accurate mathematical model, just as a human learns a new skill (such as driving, swimming, etc). Specifically, we, for the first time, propose to leverage emerging Deep Reinforcement Learning (DRL) for enabling model-free control in communication networks; and present a novel and highly effective DRL-based control framework, DRL-TE, for a fundamental networking problem: Traffic Engineering (TE). The proposed framework maximizes a widely-used utility function by jointly learning network environment and its dynamics, and making decisions under the guidance of powerful Deep Neural Networks (DNNs). We propose two new techniques, TE-aware exploration and actor-critic-based prioritized experience replay, to optimize the general DRL framework particularly for TE. To validate and evaluate the proposed framework, we implemented it in ns-3, and tested it comprehensively with both representative and randomly generated network topologies. Extensive packet-level simulation results show that 1) compared to several widely-used baseline methods, DRL-TE significantly reduces end-to-end delay and consistently improves the network utility, while offering better or comparable throughput; 2) DRL-TE is robust to network changes; and 3) DRL-TE consistently outperforms a state-ofthe-art DRL method (for continuous control), Deep Deterministic Policy Gradient (DDPG), which, however, does not offer satisfying performance. <|reference_end|>" ]

[ 7, 14, 21, 30 ]

[ "<|reference_start|> Characterization and Greedy Learning of Interventional Markov Equivalence Classes of Directed Acyclic Graphs: The investigation of directed acyclic graphs (DAGs) encoding the same Markov property, that is the same conditional independence relations of multivariate observational distributions, has a long tradition; many algorithms exist for model selection and structure learning in Markov equivalence classes. In this paper, we extend the notion of Markov equivalence of DAGs to the case of interventional distributions arising from multiple intervention experiments. We show that under reasonable assumptions on the intervention experiments, interventional Markov equivalence defines a finer partitioning of DAGs than observational Markov equivalence and hence improves the identifiability of causal models. We give a graph theoretic criterion for two DAGs being Markov equivalent under interventions and show that each interventional Markov equivalence class can, analogously to the observational case, be uniquely represented by a chain graph called interventional essential graph (also known as CPDAG in the observational case). These are key insights for deriving a generalization of the Greedy Equivalence Search algorithm aimed at structure learning from interventional data. This new algorithm is evaluated in a simulation study. <|reference_end|>", "<|reference_start|> Causal learning via manifold regularization: This paper frames causal structure estimation as a machine learning task. The idea is to treat indicators of causal relationships between variables as ‘labels’ and to exploit available data on the variables of interest to provide features for the labelling task. Background scientific knowledge or any available interventional data provide labels on some causal relationships and the remainder are treated as unlabelled. To illustrate the key ideas, we develop a distance-based approach (based on bivariate histograms) within a manifold regularization framework. We present empirical results on three different biological data sets (including examples where causal effects can be verified by experimental intervention), that together demonstrate the efficacy and general nature of the approach as well as its simplicity from a user’s point of view. <|reference_end|>", "<|reference_start|> Towards a learning theory of cause-effect inference: The first step towards the deployment of our learning setup is to guarantee the existence of a measure on the space μk(P)⇥L, where μk(P) = {μk(P ) : P 2 P} ✓ Hk is the set of kernel mean embeddings associated with the measures in P . The following lemma provides such guarantee. This allows the analysis within the rest of this Section on μk(P)⇥ L. Lemma 2. Let (Z, ⌧Z) and (L, ⌧L) be two separable topological spaces. Let P be the set of all Borel probability measures on (Z,B(⌧Z)). Let μk(P) = {μk(P ) : P 2 P} ✓ Hk, where μk is the kernel mean embedding (1) associated to some bounded continuous kernel function k : Z ⇥ Z ! R. Then, there exists a measure on μk(P)⇥ L. <|reference_end|>", "<|reference_start|> Evaluation of causal structure learning algorithms via risk estimation: Recent years have seen many advances in methods for causal structure learning from data. The empirical assessment of such methods, however, is much less developed. Motivated by this gap, we pose the following question: how can one assess, in a given problem setting, the practical efficacy of one or more causal structure learning methods? We formalize the problem in a decision-theoretic framework, via a notion of expected loss or risk for the causal setting. We introduce a theoretical notion of causal risk as well as sample quantities that can be computed from data, and study the relationship between the two, both theoretically and through an extensive simulation study. Our results provide an assumptions-light framework for assessing causal structure learning methods that can be applied in a range of practical use-cases. <|reference_end|>" ]

[ 3, 6, 7, 11 ]

[ "<|reference_start|> High-Resolution Image Synthesis with Latent Diffusion Models: By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond. Additionally, their formulation allows for a guiding mechanism to control the image generation process without retraining. However, since these models typically operate directly in pixel space, optimization of powerful DMs often consumes hundreds of GPU days and inference is expensive due to sequential evaluations. To enable DM training on limited computational resources while retaining their quality and flexibility, we apply them in the latent space of powerful pretrained autoencoders. In contrast to previous work, training diffusion models on such a representation allows for the first time to reach a near-optimal point between complexity reduction and detail preservation, greatly boosting visual fidelity. By introducing cross-attention layers into the model architecture, we turn diffusion models into powerful and flexible generators for general conditioning inputs such as text or bounding boxes and high-resolution synthesis becomes possible in a convolutional manner. Our latent diffusion models (LDMs) achieve a new state of the art for image inpainting and highly competitive performance on various tasks, including unconditional image generation, semantic scene synthesis, and super-resolution, while significantly reducing computational requirements compared to pixel-based DMs. Code is available at https://github.com/CompVis/latent-diffusion . <|reference_end|>", "<|reference_start|> Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images: We present a hierarchical VAE that, for the first time, generates samples quickly while outperforming the PixelCNN in log-likelihood on all natural image benchmarks. We begin by observing that, in theory, VAEs can actually represent autoregressive models, as well as faster, better models if they exist, when made sufficiently deep. Despite this, autoregressive models have historically outperformed VAEs in log-likelihood. We test if insufficient depth explains why by scaling a VAE to greater stochastic depth than previously explored and evaluating it CIFAR-10, ImageNet, and FFHQ. In comparison to the PixelCNN, these very deep VAEs achieve higher likelihoods, use fewer parameters, generate samples thousands of times faster, and are more easily applied to high-resolution images. Qualitative studies suggest this is because the VAE learns efficient hierarchical visual representations. We release our source code and models at https://github.com/openai/vdvae. <|reference_end|>", "<|reference_start|> SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations: Guided image synthesis enables everyday users to create and edit photo-realistic images with minimum effort. The key challenge is balancing faithfulness to the user input (e.g., hand-drawn colored strokes) and realism of the synthesized image. Existing GAN-based methods attempt to achieve such balance using either conditional GANs or GAN inversions, which are challenging and often require additional training data or loss functions for individual applications. To address these issues, we introduce a new image synthesis and editing method, Stochastic Differential Editing (SDEdit), based on a diffusion model generative prior, which synthesizes realistic images by iteratively denoising through a stochastic differential equation (SDE). Given an input image with user guide of any type, SDEdit first adds noise to the input, then subsequently denoises the resulting image through the SDE prior to increase its realism. SDEdit does not require task-specific training or inversions and can naturally achieve the balance between realism and faithfulness. SDEdit significantly outperforms state-of-the-art GAN-based methods by up to 98.09% on realism and 91.72% on overall satisfaction scores, according to a human perception study, on multiple tasks, including stroke-based image synthesis and editing as well as image compositing. <|reference_end|>", "<|reference_start|> Separate visual pathways for perception and action: <|reference_end|>" ]

[ 7, 9, 13, 15 ]

[ "<|reference_start|> Towards realistic statistical models of the grid frequency: Increased share of renewable sources of energy in a power grid leads to larger deviations in grid frequency from the nominal value resulting in more challenging control and its modelling. In this paper we focus on the grid frequency for the power system of Great Britain because the large share of renewables makes it a template for other power grids in the future and because it exhibits peculiar statistical properties, such as long-term correlations in fluctuations, periodicity, bi-modality,and heavy tails in the distribution of the grid frequency. By modifications of the swing equation and the underlying noise statistics, which we justify qualitatively and quantitatively, we reproduce these peculiar statistical properties. We apply our model to realistic frequency response services and show our predictions outperform a standard swing equation model. <|reference_end|>", "<|reference_start|> Effects of inertia, load damping and dead-bands on frequency histograms and frequency control of power systems: <|reference_end|>", "<|reference_start|> Data-driven model of the power-grid frequency dynamics: The energy system is rapidly changing to accommodate the increasing number of renewable generators and the general transition towards a more sustainable future. Simultaneously, business models and market designs evolve, affecting power-grid operation and power-grid frequency. Problems raised by this ongoing transition are increasingly addressed by transdisciplinary research approaches, ranging from purely mathematical modelling to applied case studies. These approaches require a stochastic description of consumer behaviour, fluctuations by renewables, market rules, and how they influence the stability of the power-grid frequency. Here, we introduce an easy-to-use, data-driven, stochastic model for the power-grid frequency and demonstrate how it reproduces key characteristics of the observed statistics of the Continental European and British power grids. Using data analysis tools and a Fokker–Planck approach, we estimate parameters of our deterministic and stochastic model. We offer executable code and guidelines on how to use the model on any power grid for various mathematical or engineering applications. <|reference_end|>", "<|reference_start|> Deadbands, droop, and inertia impact on power system frequency distribution: Power system inertia is falling as more energy is supplied by renewable generators, and there are concerns about the frequency controls required to guarantee satisfactory system performance. The majority of research into the negative effect of low inertia has focused on poor dynamic response following major disturbances, when the transient frequency dip can become unacceptable. However, another important practical concern—keeping average frequency deviations within acceptable limits—was mainly out of the sight of the research community. In this manuscript, we present a method for finding the frequency probability density function (PDF) for a given power system. We pass from an initial stochastic dynamic model to deterministic equations for the frequency PDF, which are analyzed to uncover key system parameters influencing frequency deviations. We show that system inertia has little effect on the frequency PDF, making virtual inertia services insufficient for keeping frequency close to nominal under ambient load fluctuations. We establish that aggregate system droop and deadband width are the only parameters that have major influence on the average frequency deviations, suggesting that energy storage might be an excellent solution for tight frequency regulation. We also show that changing the governor deadband width does not significantly affect generator movement. <|reference_end|>" ]

[ 0, 1, 2, 3 ]

[ "<|reference_start|> Introduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition: We describe the CoNLL-2003 shared task: language-independent named entity recognition. We give background information on the data sets (English and German) and the evaluation method, present a general overview of the systems that have taken part in the task and discuss their performance. <|reference_end|>", "<|reference_start|> CrossWeigh: Training Named Entity Tagger from Imperfect Annotations: Everyone makes mistakes. So do human annotators when curating labels for named entity recognition (NER). Such label mistakes might hurt model training and interfere model comparison. In this study, we dive deep into one of the widely-adopted NER benchmark datasets, CoNLL03 NER. We are able to identify label mistakes in about 5.38% test sentences, which is a significant ratio considering that the state-of-the-art test F1 score is already around 93%. Therefore, we manually correct these label mistakes and form a cleaner test set. Our re-evaluation of popular models on this corrected test set leads to more accurate assessments, compared to those on the original test set. More importantly, we propose a simple yet effective framework, CrossWeigh, to handle label mistakes during NER model training. Specifically, it partitions the training data into several folds and train independent NER models to identify potential mistakes in each fold. Then it adjusts the weights of training data accordingly to train the final NER model. Extensive experiments demonstrate significant improvements of plugging various NER models into our proposed framework on three datasets. All implementations and corrected test set are available at our Github repo: https://github.com/ZihanWangKi/CrossWeigh. <|reference_end|>", "<|reference_start|> Empower Sequence Labeling with Task-Aware Neural Language Model: Linguistic sequence labeling is a general modeling approach that encompasses a variety of problems, such as part-of-speech tagging and named entity recognition. Recent advances in neural networks (NNs) make it possible to build reliable models without handcrafted features. However, in many cases, it is hard to obtain sufficient annotations to train these models. In this study, we develop a novel neural framework to extract abundant knowledge hidden in raw texts to empower the sequence labeling task. Besides word-level knowledge contained in pre-trained word embeddings, character-aware neural language models are incorporated to extract character-level knowledge. Transfer learning techniques are further adopted to mediate different components and guide the language model towards the key knowledge. Comparing to previous methods, these task-specific knowledge allows us to adopt a more concise model and conduct more efficient training. Different from most transfer learning methods, the proposed framework does not rely on any additional supervision. It extracts knowledge from self-contained order information of training sequences. Extensive experiments on benchmark datasets demonstrate the effectiveness of leveraging character-level knowledge and the efficiency of co-training. For example, on the CoNLL03 NER task, model training completes in about 6 hours on a single GPU, reaching F1 score of 91.71$\\pm$0.10 without using any extra annotation. <|reference_end|>", "<|reference_start|> Part-of-Speech Tagging from 97% to 100%: Is It Time for Some Linguistics?: <|reference_end|>" ]

[ 8, 9, 12, 20 ]

[ "<|reference_start|> An algorithm for Tracking Multiple Targets: An algorithm for tracking multiple targets in a cluttered environment is developed. The algorithm is capable of initiating tracks, accounting for false or missing reports, and processing sets of dependent reports. As each measurement is received, probabilities are calculated for the hypotheses that the measurement came from previously known targets in a target file, or from a new target, or that the measurement is false. Target states are estimated from each such data-association hypothesis, using a Kalman filter. As more measurements are received, the probabilities of joint hypotheses are calculated recursively using all available information such as density of unknown targets, density of false targets, probability of detection, and location uncertainty. This branching technique allows correlation of a measurement with its source based on subsequent, as well as previous, data. To keep the number of hypotheses reasonable, unlikely hypotheses are eliminated and hypotheses with similar target estimates are combined. To minimize computational requirements, the entire set of targets and measurements is divided into clusters that are solved independently. In an illustrative example of aircraft tracking, the algorithm successfully tracks targets over a wide range of conditions. <|reference_end|>", "<|reference_start|> Online Multiperson Tracking-by-Detection from a Single, Uncalibrated\nCamera: In this paper, we address the problem of automatically detecting and tracking a variable number of persons in complex scenes using a monocular, potentially moving, uncalibrated camera. We propose a novel approach for multiperson tracking-by-detection in a particle filtering framework. In addition to final high-confidence detections, our algorithm uses the continuous confidence of pedestrian detectors and online-trained, instance-specific classifiers as a graded observation model. Thus, generic object category knowledge is complemented by instance-specific information. The main contribution of this paper is to explore how these unreliable information sources can be used for robust multiperson tracking. The algorithm detects and tracks a large number of dynamically moving people in complex scenes with occlusions, does not rely on background modeling, requires no camera or ground plane calibration, and only makes use of information from the past. Hence, it imposes very few restrictions and is suitable for online applications. Our experiments show that the method yields good tracking performance in a large variety of highly dynamic scenarios, such as typical surveillance videos, webcam footage, or sports sequences. We demonstrate that our algorithm outperforms other methods that rely on additional information. Furthermore, we analyze the influence of different algorithm components on the robustness. <|reference_end|>", "<|reference_start|> Shape-based online multitarget tracking and detection for targets causing multiple measurements: Variational Bayesian clustering and lossless data association: This paper proposes a novel online two-level multitarget tracking and detection (MTTD) algorithm. The algorithm focuses on multitarget detection and tracking for the case of multiple measurements per target and for an unknown and varying number of targets. Information is continuously exchanged in both directions between the two levels. Using the high level target position and shape information, the low level clusters the measurements. Furthermore, the low level features automatic relevance detection (ARD), as it automatically determines the optimal number of clusters from the measurements taking into account the expected target shapes. The high level's data association allows for a varying number of targets. A joint probabilistic data association algorithm looks for associations between clusters of measurements and targets. These associations are used to update the target trackers and the target shapes with the individual measurements. No information is lost in the two-level approach since the measurement information is not summarized into features. The target trackers are based on an underlying motion model, while the high level is supplemented with a filter estimating the number of targets. The algorithm is verified using both simulations and experiments using two sensor modalities, video and laser scanner, for detection and tracking of people and ants. <|reference_end|>", "<|reference_start|> {Global data association for multi-object tracking using network flows: We propose a network flow based optimization method for data association needed for multiple object tracking. The maximum-a-posteriori (MAP) data association problem is mapped into a cost-flow network with a non-overlap constraint on trajectories. The optimal data association is found by a min-cost flow algorithm in the network. The network is augmented to include an explicit occlusion model(EOM) to track with long-term inter-object occlusions. A solution to the EOM-based network is found by an iterative approach built upon the original algorithm. Initialization and termination of trajectories and potential false observations are modeled by the formulation intrinsically. The method is efficient and does not require hypotheses pruning. Performance is compared with previous results on two public pedestrian datasets to show its improvement. <|reference_end|>" ]

[ 3, 4, 5, 7 ]