Paper Group ANR 156
Ordered Functional Decision Diagrams. On Thompson Sampling with Langevin Algorithms. Optimizing Coordinated Vehicle Platooning: An Analytical Approach Based on Stochastic Dynamic Programming. Federated Learning of a Mixture of Global and Local Models. Reliable Distributed Clustering with Redundant Data Assignment. PointNetKL: Deep Inference for GIC …
Ordered Functional Decision Diagrams
| Title | Ordered Functional Decision Diagrams |
| Authors | Joan Thibault, Khalil Ghorbal |
| Abstract | Several BDD variants were designed to exploit special features of Boolean functions to achieve better compression rates.Deciding a priori which variant to use is as hard as constructing the diagrams themselves and the conversion between variants comes in general with a prohibitive cost.This observation leads naturally to a growing interest into when and how one can combine existing variants to benefit from their respective sweet spots.In this paper, we introduce a novel framework, termed \lambdaDD (LDD), that revisits BDD from a purely functional point of view.The framework allows to classify the already existing variants, including the most recent ones like ChainDD and ESRBDD, as implementations of a special class of ordered models.We enumerate, in a principled way, all the models of this class and isolate its most expressive model.This new model, termed \lambdaDD-O-NUCX, is suitable for both dense and sparse Boolean functions, and, unlike ChainDD and ESRBDD, is invariant by negation.The canonicity of \lambdaDD-O-NUCX is formally verified using the Coq proof assistant.We furthermore provide experimental evidence corroborating our theoretical findings: more expressive \lambdaDD models achieve, indeed, better memory compression rates. |
| Tasks | |
| Published | 2020-03-20 |
| URL | https://arxiv.org/abs/2003.09340v1 |
| https://arxiv.org/pdf/2003.09340v1.pdf | |
| PWC | https://paperswithcode.com/paper/ordered-functional-decision-diagrams |
| Repo | |
| Framework | |
On Thompson Sampling with Langevin Algorithms
| Title | On Thompson Sampling with Langevin Algorithms |
| Authors | Eric Mazumdar, Aldo Pacchiano, Yi-an Ma, Peter L. Bartlett, Michael I. Jordan |
| Abstract | Thompson sampling is a methodology for multi-armed bandit problems that is known to enjoy favorable performance in both theory and practice. It does, however, have a significant limitation computationally, arising from the need for samples from posterior distributions at every iteration. We propose two Markov Chain Monte Carlo (MCMC) methods tailored to Thompson sampling to address this issue. We construct quickly converging Langevin algorithms to generate approximate samples that have accuracy guarantees, and we leverage novel posterior concentration rates to analyze the regret of the resulting approximate Thompson sampling algorithm. Further, we specify the necessary hyper-parameters for the MCMC procedure to guarantee optimal instance-dependent frequentist regret while having low computational complexity. In particular, our algorithms take advantage of both posterior concentration and a sample reuse mechanism to ensure that only a constant number of iterations and a constant amount of data is needed in each round. The resulting approximate Thompson sampling algorithm has logarithmic regret and its computational complexity does not scale with the time horizon of the algorithm. |
| Tasks | |
| Published | 2020-02-23 |
| URL | https://arxiv.org/abs/2002.10002v1 |
| https://arxiv.org/pdf/2002.10002v1.pdf | |
| PWC | https://paperswithcode.com/paper/on-thompson-sampling-with-langevin-algorithms |
| Repo | |
| Framework | |
Optimizing Coordinated Vehicle Platooning: An Analytical Approach Based on Stochastic Dynamic Programming
| Title | Optimizing Coordinated Vehicle Platooning: An Analytical Approach Based on Stochastic Dynamic Programming |
| Authors | Xi Xiong, Junyi Sha, Li Jin |
| Abstract | Platooning connected and autonomous vehicles (CAVs) can improve traffic and fuel efficiency. However, scalable platooning operations requires junction-level coordination, which has not been well studied. In this paper, we study the coordination of vehicle platooning at highway junctions. We consider a setting where CAVs randomly arrive at a highway junction according to a general renewal process. When a CAV approaches the junction, a system operator determines whether the CAV will merge into the platoon ahead according to the positions and speeds of the CAV and the platoon. We formulate a Markov decision process to minimize the discounted cumulative travel cost, i.e. fuel consumption plus travel delay, over an infinite time horizon. We show that the optimal policy is threshold-based: the CAV will merge with the platoon if and only if the difference between the CAV’s and the platoon’s predicted times of arrival at the junction is less than a constant threshold. We also propose two ready-to-implement algorithms to derive the optimal policy. Comparison with the classical value iteration algorithm implies that our approach explicitly incorporating the characteristics of the optimal policy is significantly more efficient in terms of computation. Importantly, we show that the optimal policy under Poisson arrivals can be obtained by solving a system of integral equations. We also validate our results in simulation with Real-time Strategy (RTS) using real traffic data. The simulation results indicate that the proposed method yields better performance compared with the conventional method. |
| Tasks | Autonomous Vehicles |
| Published | 2020-03-29 |
| URL | https://arxiv.org/abs/2003.13067v1 |
| https://arxiv.org/pdf/2003.13067v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimizing-coordinated-vehicle-platooning-an |
| Repo | |
| Framework | |
Federated Learning of a Mixture of Global and Local Models
| Title | Federated Learning of a Mixture of Global and Local Models |
| Authors | Filip Hanzely, Peter Richtárik |
| Abstract | We propose a new optimization formulation for training federated learning models. The standard formulation has the form of an empirical risk minimization problem constructed to find a single global model trained from the private data stored across all participating devices. In contrast, our formulation seeks an explicit trade-off between this traditional global model and the local models, which can be learned by each device from its own private data without any communication. Further, we develop several efficient variants of SGD (with and without partial participation and with and without variance reduction) for solving the new formulation and prove communication complexity guarantees. Notably, our methods are similar but not identical to federated averaging / local SGD, thus shedding some light on the essence of the elusive method. In particular, our methods do not perform full averaging steps and instead merely take steps towards averaging. We argue for the benefits of this new paradigm for federated learning. |
| Tasks | |
| Published | 2020-02-10 |
| URL | https://arxiv.org/abs/2002.05516v2 |
| https://arxiv.org/pdf/2002.05516v2.pdf | |
| PWC | https://paperswithcode.com/paper/federated-learning-of-a-mixture-of-global-and |
| Repo | |
| Framework | |
Reliable Distributed Clustering with Redundant Data Assignment
| Title | Reliable Distributed Clustering with Redundant Data Assignment |
| Authors | Venkata Gandikota, Arya Mazumdar, Ankit Singh Rawat |
| Abstract | In this paper, we present distributed generalized clustering algorithms that can handle large scale data across multiple machines in spite of straggling or unreliable machines. We propose a novel data assignment scheme that enables us to obtain global information about the entire data even when some machines fail to respond with the results of the assigned local computations. The assignment scheme leads to distributed algorithms with good approximation guarantees for a variety of clustering and dimensionality reduction problems. |
| Tasks | Dimensionality Reduction |
| Published | 2020-02-20 |
| URL | https://arxiv.org/abs/2002.08892v1 |
| https://arxiv.org/pdf/2002.08892v1.pdf | |
| PWC | https://paperswithcode.com/paper/reliable-distributed-clustering-with |
| Repo | |
| Framework | |
PointNetKL: Deep Inference for GICP Covariance Estimation in Bathymetric SLAM
| Title | PointNetKL: Deep Inference for GICP Covariance Estimation in Bathymetric SLAM |
| Authors | Ignacio Torroba, Christopher Iliffe Sprague, Nils Bore, John Folkesson |
| Abstract | Registration methods for point clouds have become a key component of many SLAM systems on autonomous vehicles. However, an accurate estimate of the uncertainty of such registration is a key requirement to a consistent fusion of this kind of measurements in a SLAM filter. This estimate, which is normally given as a covariance in the transformation computed between point cloud reference frames, has been modelled following different approaches, among which the most accurate is considered to be the Monte Carlo method. However, a Monte Carlo approximation is cumbersome to use inside a time-critical application such as online SLAM. Efforts have been made to estimate this covariance via machine learning using carefully designed features to abstract the raw point clouds. However, the performance of this approach is sensitive to the features chosen. We argue that it is possible to learn the features along with the covariance by working with the raw data and thus we propose a new approach based on PointNet. In this work, we train this network using the KL divergence between the learned uncertainty distribution and one computed by the Monte Carlo method as the loss. We test the performance of the general model presented applying it to our target use-case of SLAM with an autonomous underwater vehicle (AUV) restricted to the 2-dimensional registration of 3D bathymetric point clouds. |
| Tasks | Autonomous Vehicles |
| Published | 2020-03-24 |
| URL | https://arxiv.org/abs/2003.10931v1 |
| https://arxiv.org/pdf/2003.10931v1.pdf | |
| PWC | https://paperswithcode.com/paper/pointnetkl-deep-inference-for-gicp-covariance |
| Repo | |
| Framework | |
Using word embeddings to improve the discriminability of co-occurrence text networks
| Title | Using word embeddings to improve the discriminability of co-occurrence text networks |
| Authors | Laura V. C. Quispe, Jorge A. V. Tohalino, Diego R. Amancio |
| Abstract | Word co-occurrence networks have been employed to analyze texts both in the practical and theoretical scenarios. Despite the relative success in several applications, traditional co-occurrence networks fail in establishing links between similar words whenever they appear distant in the text. Here we investigate whether the use of word embeddings as a tool to create virtual links in co-occurrence networks may improve the quality of classification systems. Our results revealed that the discriminability in the stylometry task is improved when using Glove, Word2Vec and FastText. In addition, we found that optimized results are obtained when stopwords are not disregarded and a simple global thresholding strategy is used to establish virtual links. Because the proposed approach is able to improve the representation of texts as complex networks, we believe that it could be extended to study other natural language processing tasks. Likewise, theoretical languages studies could benefit from the adopted enriched representation of word co-occurrence networks. |
| Tasks | Word Embeddings |
| Published | 2020-03-13 |
| URL | https://arxiv.org/abs/2003.06279v1 |
| https://arxiv.org/pdf/2003.06279v1.pdf | |
| PWC | https://paperswithcode.com/paper/using-word-embeddings-to-improve-the |
| Repo | |
| Framework | |
A Neural Architecture for Detecting Confusion in Eye-tracking Data
| Title | A Neural Architecture for Detecting Confusion in Eye-tracking Data |
| Authors | Shane Sims, Cristina Conati |
| Abstract | Encouraged by the success of deep learning in a variety of domains, we investigate a novel application of its methods on the effectiveness of detecting user confusion in eye-tracking data. We introduce an architecture that uses RNN and CNN sub-models in parallel to take advantage of the temporal and visuospatial aspects of our data. Experiments with a dataset of user interactions with the ValueChart visualization tool show that our model outperforms an existing model based on Random Forests resulting in a 22% improvement in combined sensitivity & specificity. |
| Tasks | Eye Tracking |
| Published | 2020-03-13 |
| URL | https://arxiv.org/abs/2003.06434v1 |
| https://arxiv.org/pdf/2003.06434v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-neural-architecture-for-detecting-confusion |
| Repo | |
| Framework | |
Softmax-based Classification is k-means Clustering: Formal Proof, Consequences for Adversarial Attacks, and Improvement through Centroid Based Tailoring
| Title | Softmax-based Classification is k-means Clustering: Formal Proof, Consequences for Adversarial Attacks, and Improvement through Centroid Based Tailoring |
| Authors | Sibylle Hess, Wouter Duivesteijn, Decebal Mocanu |
| Abstract | We formally prove the connection between k-means clustering and the predictions of neural networks based on the softmax activation layer. In existing work, this connection has been analyzed empirically, but it has never before been mathematically derived. The softmax function partitions the transformed input space into cones, each of which encompasses a class. This is equivalent to putting a number of centroids in this transformed space at equal distance from the origin, and k-means clustering the data points by proximity to these centroids. Softmax only cares in which cone a data point falls, and not how far from the centroid it is within that cone. We formally prove that networks with a small Lipschitz modulus (which corresponds to a low susceptibility to adversarial attacks) map data points closer to the cluster centroids, which results in a mapping to a k-means-friendly space. To leverage this knowledge, we propose Centroid Based Tailoring as an alternative to the softmax function in the last layer of a neural network. The resulting Gauss network has similar predictive accuracy as traditional networks, but is less susceptible to one-pixel attacks; while the main contribution of this paper is theoretical in nature, the Gauss network contributes empirical auxiliary benefits. |
| Tasks | |
| Published | 2020-01-07 |
| URL | https://arxiv.org/abs/2001.01987v1 |
| https://arxiv.org/pdf/2001.01987v1.pdf | |
| PWC | https://paperswithcode.com/paper/softmax-based-classification-is-k-means |
| Repo | |
| Framework | |
A new geodesic-based feature for characterization of 3D shapes: application to soft tissue organ temporal deformations
| Title | A new geodesic-based feature for characterization of 3D shapes: application to soft tissue organ temporal deformations |
| Authors | Karim Makki, Amine Bohi, Augustin C. Ogier, Marc-Emmanuel Bellemare |
| Abstract | In this paper, we propose a method for characterizing 3D shapes from point clouds and we show a direct application on a study of organ temporal deformations. As an example, we characterize the behavior of a bladder during a forced respiratory motion with a reduced number of 3D surface points: first, a set of equidistant points representing the vertices of quadrilateral mesh for the surface in the first time frame are tracked throughout a long dynamic MRI sequence using a Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework. Second, a novel geometric feature which is invariant to scaling and rotation is proposed for characterizing the temporal organ deformations by employing an Eulerian Partial Differential Equations (PDEs) methodology. We demonstrate the robustness of our feature on both synthetic 3D shapes and realistic dynamic MRI data portraying the bladder deformation during forced respiratory motions. Promising results are obtained, showing that the proposed feature may be useful for several computer vision applications such as medical imaging, aerodynamics and robotics. |
| Tasks | |
| Published | 2020-03-18 |
| URL | https://arxiv.org/abs/2003.08332v1 |
| https://arxiv.org/pdf/2003.08332v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-new-geodesic-based-feature-for |
| Repo | |
| Framework | |
A Hybrid Quantum enabled RBM Advantage: Convolutional Autoencoders For Quantum Image Compression and Generative Learning
| Title | A Hybrid Quantum enabled RBM Advantage: Convolutional Autoencoders For Quantum Image Compression and Generative Learning |
| Authors | Jennifer Sleeman, John Dorband, Milton Halem |
| Abstract | Understanding how the D-Wave quantum computer could be used for machine learning problems is of growing interest. Our work evaluates the feasibility of using the D-Wave as a sampler for machine learning. We describe a hybrid system that combines a classical deep neural network autoencoder with a quantum annealing Restricted Boltzmann Machine (RBM) using the D-Wave. We evaluate our hybrid autoencoder algorithm using two datasets, the MNIST dataset and MNIST Fashion dataset. We evaluate the quality of this method by using a downstream classification method where the training is based on quantum RBM-generated samples. Our method overcomes two key limitations in the current 2000-qubit D-Wave processor, namely the limited number of qubits available to accommodate typical problem sizes for fully connected quantum objective functions and samples that are binary pixel representations. As a consequence of these limitations we are able to show how we achieved nearly a 22-fold compression factor of grayscale 28 x 28 sized images to binary 6 x 6 sized images with a lossy recovery of the original 28 x 28 grayscale images. We further show how generating samples from the D-Wave after training the RBM, resulted in 28 x 28 images that were variations of the original input data distribution, as opposed to recreating the training samples. We formulated an MNIST classification problem using a deep convolutional neural network that used samples from a quantum RBM to train the MNIST classifier and compared the results with an MNIST classifier trained with the original MNIST training data set, as well as an MNIST classifier trained using classical RBM samples. Our hybrid autoencoder approach indicates advantage for RBM results relative to the use of a current RBM classical computer implementation for image-based machine learning and even more promising results for the next generation D-Wave quantum system. |
| Tasks | Image Compression |
| Published | 2020-01-31 |
| URL | https://arxiv.org/abs/2001.11946v1 |
| https://arxiv.org/pdf/2001.11946v1.pdf | |
| PWC | https://paperswithcode.com/paper/a-hybrid-quantum-enabled-rbm-advantage |
| Repo | |
| Framework | |
Audio-video Emotion Recognition in the Wild using Deep Hybrid Networks
| Title | Audio-video Emotion Recognition in the Wild using Deep Hybrid Networks |
| Authors | Xin Guo, Luisa F. Polanía, Kenneth E. Barner |
| Abstract | This paper presents an audiovisual-based emotion recognition hybrid network. While most of the previous work focuses either on using deep models or hand-engineered features extracted from images, we explore multiple deep models built on both images and audio signals. Specifically, in addition to convolutional neural networks (CNN) and recurrent neutral networks (RNN) trained on facial images, the hybrid network also contains one SVM classifier trained on holistic acoustic feature vectors, one long short-term memory network (LSTM) trained on short-term feature sequences extracted from segmented audio clips, and one Inception(v2)-LSTM network trained on image-like maps, which are built based on short-term acoustic feature sequences. Experimental results show that the proposed hybrid network outperforms the baseline method by a large margin. |
| Tasks | Emotion Recognition, Video Emotion Recognition |
| Published | 2020-02-20 |
| URL | https://arxiv.org/abs/2002.09023v1 |
| https://arxiv.org/pdf/2002.09023v1.pdf | |
| PWC | https://paperswithcode.com/paper/audio-video-emotion-recognition-in-the-wild |
| Repo | |
| Framework | |
Dual-attention Guided Dropblock Module for Weakly Supervised Object Localization
| Title | Dual-attention Guided Dropblock Module for Weakly Supervised Object Localization |
| Authors | Junhui Yin, Siqing Zhang, Dongliang Chang, Zhanyu Ma, Jun Guo |
| Abstract | In this paper, we present a dual-attention guided dropblock module, and aim at learning the informative and complementary visual features for weakly supervised object localization (WSOL). The attention mechanism is extended to the task of WSOL, and design two types of attention modules to learn the discriminative features for better feature representations. Based on two types of attention mechanism, we propose a channel attention guided dropout (CAGD) and a spatial attention guided dropblock (SAGD). The CAGD ranks channel attention by a measure of importance and consider the top-k largest magnitude attentions as important ones. The SAGD can not only completely remove the information by erasing the contiguous regions of feature maps rather than individual pixels, but also simply distinguish the foreground objects and background regions to alleviate the attention misdirection. Extensive experiments demonstrate that the proposed method achieves new state-of-the-art localization accuracy on a challenging dataset. |
| Tasks | Object Localization, Weakly-Supervised Object Localization |
| Published | 2020-03-09 |
| URL | https://arxiv.org/abs/2003.04719v2 |
| https://arxiv.org/pdf/2003.04719v2.pdf | |
| PWC | https://paperswithcode.com/paper/dual-attention-guided-dropblock-module-for |
| Repo | |
| Framework | |
Modeling Information Need of Users in Search Sessions
| Title | Modeling Information Need of Users in Search Sessions |
| Authors | Kishaloy Halder, Heng-Tze Cheng, Ellie Ka In Chio, Georgios Roumpos, Tao Wu, Ritesh Agarwal |
| Abstract | Users issue queries to Search Engines, and try to find the desired information in the results produced. They repeat this process if their information need is not met at the first place. It is crucial to identify the important words in a query that depict the actual information need of the user and will determine the course of a search session. To this end, we propose a sequence-to-sequence based neural architecture that leverages the set of past queries issued by users, and results that were explored by them. Firstly, we employ our model for predicting the words in the current query that are important and would be retained in the next query. Additionally, as a downstream application of our model, we evaluate it on the widely popular task of next query suggestion. We show that our intuitive strategy of capturing information need can yield superior performance at these tasks on two large real-world search log datasets. |
| Tasks | |
| Published | 2020-01-03 |
| URL | https://arxiv.org/abs/2001.00861v1 |
| https://arxiv.org/pdf/2001.00861v1.pdf | |
| PWC | https://paperswithcode.com/paper/modeling-information-need-of-users-in-search |
| Repo | |
| Framework | |
Optimizing Wireless Systems Using Unsupervised and Reinforced-Unsupervised Deep Learning
| Title | Optimizing Wireless Systems Using Unsupervised and Reinforced-Unsupervised Deep Learning |
| Authors | Dong Liu, Chengjian Sun, Chenyang Yang, Lajos Hanzo |
| Abstract | Resource allocation and transceivers in wireless networks are usually designed by solving optimization problems subject to specific constraints, which can be formulated as variable or functional optimization. If the objective and constraint functions of a variable optimization problem can be derived, standard numerical algorithms can be applied for finding the optimal solution, which however incur high computational cost when the dimension of the variable is high. To reduce the on-line computational complexity, learning the optimal solution as a function of the environment’s status by deep neural networks (DNNs) is an effective approach. DNNs can be trained under the supervision of optimal solutions, which however, is not applicable to the scenarios without models or for functional optimization where the optimal solutions are hard to obtain. If the objective and constraint functions are unavailable, reinforcement learning can be applied to find the solution of a functional optimization problem, which is however not tailored to optimization problems in wireless networks. In this article, we introduce unsupervised and reinforced-unsupervised learning frameworks for solving both variable and functional optimization problems without the supervision of the optimal solutions. When the mathematical model of the environment is completely known and the distribution of environment’s status is known or unknown, we can invoke unsupervised learning algorithm. When the mathematical model of the environment is incomplete, we introduce reinforced-unsupervised learning algorithms that learn the model by interacting with the environment. Our simulation results confirm the applicability of these learning frameworks by taking a user association problem as an example. |
| Tasks | |
| Published | 2020-01-03 |
| URL | https://arxiv.org/abs/2001.00784v1 |
| https://arxiv.org/pdf/2001.00784v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimizing-wireless-systems-using |
| Repo | |
| Framework | |
