May 6, 2019

2896 words 14 mins read

Paper Group ANR 291

Discriminative Sparsity for Sonar ATR. Convex block-sparse linear regression with expanders – provably. FaceVR: Real-Time Facial Reenactment and Eye Gaze Control in Virtual Reality. Ensemble Robustness and Generalization of Stochastic Deep Learning Algorithms. Higher-Order Low-Rank Regression. Learning Temporal Transformations From Time-Lapse Vide …

Discriminative Sparsity for Sonar ATR


Title	Discriminative Sparsity for Sonar ATR
Authors	John McKay, Raghu Raj, Vishal Monga, Jason Isaacs
Abstract	Advancements in Sonar image capture have enabled researchers to apply sophisticated object identification algorithms in order to locate targets of interest in images such as mines. Despite progress in this field, modern sonar automatic target recognition (ATR) approaches lack robustness to the amount of noise one would expect in real-world scenarios, the capability to handle blurring incurred from the physics of image capture, and the ability to excel with relatively few training samples. We address these challenges by adapting modern sparsity-based techniques with dictionaries comprising of training from each class. We develop new discriminative (as opposed to generative) sparse representations which can help automatically classify targets in Sonar imaging. Using a simulated SAS data set from the Naval Surface Warfare Center (NSWC), we obtained compelling classification rates for multi-class problems even in cases with considerable noise and sparsity in training samples.
Tasks
Published	2016-01-01
URL	http://arxiv.org/abs/1601.00119v1
PDF	http://arxiv.org/pdf/1601.00119v1.pdf
PWC	https://paperswithcode.com/paper/discriminative-sparsity-for-sonar-atr
Repo
Framework

Convex block-sparse linear regression with expanders – provably


Title	Convex block-sparse linear regression with expanders – provably
Authors	Anastasios Kyrillidis, Bubacarr Bah, Rouzbeh Hasheminezhad, Quoc Tran-Dinh, Luca Baldassarre, Volkan Cevher
Abstract	Sparse matrices are favorable objects in machine learning and optimization. When such matrices are used, in place of dense ones, the overall complexity requirements in optimization can be significantly reduced in practice, both in terms of space and run-time. Prompted by this observation, we study a convex optimization scheme for block-sparse recovery from linear measurements. To obtain linear sketches, we use expander matrices, i.e., sparse matrices containing only few non-zeros per column. Hitherto, to the best of our knowledge, such algorithmic solutions have been only studied from a non-convex perspective. Our aim here is to theoretically characterize the performance of convex approaches under such setting. Our key novelty is the expression of the recovery error in terms of the model-based norm, while assuring that solution lives in the model. To achieve this, we show that sparse model-based matrices satisfy a group version of the null-space property. Our experimental findings on synthetic and real applications support our claims for faster recovery in the convex setting – as opposed to using dense sensing matrices, while showing a competitive recovery performance.
Tasks
Published	2016-03-21
URL	http://arxiv.org/abs/1603.06313v2
PDF	http://arxiv.org/pdf/1603.06313v2.pdf
PWC	https://paperswithcode.com/paper/convex-block-sparse-linear-regression-with
Repo
Framework

FaceVR: Real-Time Facial Reenactment and Eye Gaze Control in Virtual Reality


Title	FaceVR: Real-Time Facial Reenactment and Eye Gaze Control in Virtual Reality
Authors	Justus Thies, Michael Zollhöfer, Marc Stamminger, Christian Theobalt, Matthias Nießner
Abstract	We propose FaceVR, a novel image-based method that enables video teleconferencing in VR based on self-reenactment. State-of-the-art face tracking methods in the VR context are focused on the animation of rigged 3d avatars. While they achieve good tracking performance the results look cartoonish and not real. In contrast to these model-based approaches, FaceVR enables VR teleconferencing using an image-based technique that results in nearly photo-realistic outputs. The key component of FaceVR is a robust algorithm to perform real-time facial motion capture of an actor who is wearing a head-mounted display (HMD), as well as a new data-driven approach for eye tracking from monocular videos. Based on reenactment of a prerecorded stereo video of the person without the HMD, FaceVR incorporates photo-realistic re-rendering in real time, thus allowing artificial modifications of face and eye appearances. For instance, we can alter facial expressions or change gaze directions in the prerecorded target video. In a live setup, we apply these newly-introduced algorithmic components.
Tasks	Eye Tracking, Motion Capture
Published	2016-10-11
URL	http://arxiv.org/abs/1610.03151v2
PDF	http://arxiv.org/pdf/1610.03151v2.pdf
PWC	https://paperswithcode.com/paper/facevr-real-time-facial-reenactment-and-eye
Repo
Framework

Ensemble Robustness and Generalization of Stochastic Deep Learning Algorithms


Title	Ensemble Robustness and Generalization of Stochastic Deep Learning Algorithms
Authors	Tom Zahavy, Bingyi Kang, Alex Sivak, Jiashi Feng, Huan Xu, Shie Mannor
Abstract	The question why deep learning algorithms generalize so well has attracted increasing research interest. However, most of the well-established approaches, such as hypothesis capacity, stability or sparseness, have not provided complete explanations (Zhang et al., 2016; Kawaguchi et al., 2017). In this work, we focus on the robustness approach (Xu & Mannor, 2012), i.e., if the error of a hypothesis will not change much due to perturbations of its training examples, then it will also generalize well. As most deep learning algorithms are stochastic (e.g., Stochastic Gradient Descent, Dropout, and Bayes-by-backprop), we revisit the robustness arguments of Xu & Mannor, and introduce a new approach, ensemble robustness, that concerns the robustness of a population of hypotheses. Through the lens of ensemble robustness, we reveal that a stochastic learning algorithm can generalize well as long as its sensitiveness to adversarial perturbations is bounded in average over training examples. Moreover, an algorithm may be sensitive to some adversarial examples (Goodfellow et al., 2015) but still generalize well. To support our claims, we provide extensive simulations for different deep learning algorithms and different network architectures exhibiting a strong correlation between ensemble robustness and the ability to generalize.
Tasks
Published	2016-02-07
URL	http://arxiv.org/abs/1602.02389v4
PDF	http://arxiv.org/pdf/1602.02389v4.pdf
PWC	https://paperswithcode.com/paper/ensemble-robustness-and-generalization-of
Repo
Framework

Higher-Order Low-Rank Regression


Title	Higher-Order Low-Rank Regression
Authors	Guillaume Rabusseau, Hachem Kadri
Abstract	This paper proposes an efficient algorithm (HOLRR) to handle regression tasks where the outputs have a tensor structure. We formulate the regression problem as the minimization of a least square criterion under a multilinear rank constraint, a difficult non convex problem. HOLRR computes efficiently an approximate solution of this problem, with solid theoretical guarantees. A kernel extension is also presented. Experiments on synthetic and real data show that HOLRR outperforms multivariate and multilinear regression methods and is considerably faster than existing tensor methods.
Tasks
Published	2016-02-22
URL	http://arxiv.org/abs/1602.06863v1
PDF	http://arxiv.org/pdf/1602.06863v1.pdf
PWC	https://paperswithcode.com/paper/higher-order-low-rank-regression
Repo
Framework

Learning Temporal Transformations From Time-Lapse Videos


Title	Learning Temporal Transformations From Time-Lapse Videos
Authors	Yipin Zhou, Tamara L. Berg
Abstract	Based on life-long observations of physical, chemical, and biologic phenomena in the natural world, humans can often easily picture in their minds what an object will look like in the future. But, what about computers? In this paper, we learn computational models of object transformations from time-lapse videos. In particular, we explore the use of generative models to create depictions of objects at future times. These models explore several different prediction tasks: generating a future state given a single depiction of an object, generating a future state given two depictions of an object at different times, and generating future states recursively in a recurrent framework. We provide both qualitative and quantitative evaluations of the generated results, and also conduct a human evaluation to compare variations of our models.
Tasks
Published	2016-08-27
URL	http://arxiv.org/abs/1608.07724v1
PDF	http://arxiv.org/pdf/1608.07724v1.pdf
PWC	https://paperswithcode.com/paper/learning-temporal-transformations-from-time
Repo
Framework

A Non-generative Framework and Convex Relaxations for Unsupervised Learning


Title	A Non-generative Framework and Convex Relaxations for Unsupervised Learning
Authors	Elad Hazan, Tengyu Ma
Abstract	We give a novel formal theoretical framework for unsupervised learning with two distinctive characteristics. First, it does not assume any generative model and based on a worst-case performance metric. Second, it is comparative, namely performance is measured with respect to a given hypothesis class. This allows to avoid known computational hardness results and improper algorithms based on convex relaxations. We show how several families of unsupervised learning models, which were previously only analyzed under probabilistic assumptions and are otherwise provably intractable, can be efficiently learned in our framework by convex optimization.
Tasks
Published	2016-10-04
URL	http://arxiv.org/abs/1610.01132v3
PDF	http://arxiv.org/pdf/1610.01132v3.pdf
PWC	https://paperswithcode.com/paper/a-non-generative-framework-and-convex
Repo
Framework

Exploring Local Context for Multi-target Tracking in Wide Area Aerial Surveillance


Title	Exploring Local Context for Multi-target Tracking in Wide Area Aerial Surveillance
Authors	Bor-Jeng Chen, Gerard Medioni
Abstract	Tracking many vehicles in wide coverage aerial imagery is crucial for understanding events in a large field of view. Most approaches aim to associate detections from frame differencing into tracks. However, slow or stopped vehicles result in long-term missing detections and further cause tracking discontinuities. Relying merely on appearance clue to recover missing detections is difficult as targets are extremely small and in grayscale. In this paper, we address the limitations of detection association methods by coupling it with a local context tracker (LCT), which does not rely on motion detections. On one hand, our LCT learns neighboring spatial relation and tracks each target in consecutive frames using graph optimization. It takes the advantage of context constraints to avoid drifting to nearby targets. We generate hypotheses from sparse and dense flow efficiently to keep solutions tractable. On the other hand, we use detection association strategy to extract short tracks in batch processing. We explicitly handle merged detections by generating additional hypotheses from them. Our evaluation on wide area aerial imagery sequences shows significant improvement over state-of-the-art methods.
Tasks
Published	2016-03-28
URL	http://arxiv.org/abs/1603.08592v1
PDF	http://arxiv.org/pdf/1603.08592v1.pdf
PWC	https://paperswithcode.com/paper/exploring-local-context-for-multi-target
Repo
Framework

Simultaneous Safe Screening of Features and Samples in Doubly Sparse Modeling


Title	Simultaneous Safe Screening of Features and Samples in Doubly Sparse Modeling
Authors	Atsushi Shibagaki, Masayuki Karasuyama, Kohei Hatano, Ichiro Takeuchi
Abstract	The problem of learning a sparse model is conceptually interpreted as the process of identifying active features/samples and then optimizing the model over them. Recently introduced safe screening allows us to identify a part of non-active features/samples. So far, safe screening has been individually studied either for feature screening or for sample screening. In this paper, we introduce a new approach for safely screening features and samples simultaneously by alternatively iterating feature and sample screening steps. A significant advantage of considering them simultaneously rather than individually is that they have a synergy effect in the sense that the results of the previous safe feature screening can be exploited for improving the next safe sample screening performances, and vice-versa. We first theoretically investigate the synergy effect, and then illustrate the practical advantage through intensive numerical experiments for problems with large numbers of features and samples.
Tasks
Published	2016-02-08
URL	http://arxiv.org/abs/1602.02485v1
PDF	http://arxiv.org/pdf/1602.02485v1.pdf
PWC	https://paperswithcode.com/paper/simultaneous-safe-screening-of-features-and
Repo
Framework

Extraction of clinical information from the non-invasive fetal electrocardiogram


Title	Extraction of clinical information from the non-invasive fetal electrocardiogram
Authors	Joachim Behar
Abstract	Estimation of the fetal heart rate (FHR) has gained interest in the last century, low heart rate variability has been studied to identify intrauterine growth restricted fetuses (prepartum), and abnormal FHR patterns have been associated with fetal distress during delivery (intrapartum). Several monitoring techniques have been proposed for FHR estimation, including auscultation and Doppler ultrasound. This thesis focuses on the extraction of the non-invasive fetal electrocardiogram (NI-FECG) recorded from a limited set of abdominal sensors. The main challenge with NI-FECG extraction techniques is the low signal-to-noise ratio of the FECG signal on the abdominal mixture signal which consists of a dominant maternal ECG component, FECG and noise. However the NI-FECG offers many advantages over the alternative fetal monitoring techniques, the most important one being the opportunity to enable morphological analysis of the FECG which is vital for determining whether an observed FHR event is normal or pathological. In order to advance the field of NI-FECG signal processing, the development of standardised public databases and benchmarking of a number of published and novel algorithms was necessary.
Tasks	Heart Rate Variability, Morphological Analysis
Published	2016-05-27
URL	http://arxiv.org/abs/1606.01093v1
PDF	http://arxiv.org/pdf/1606.01093v1.pdf
PWC	https://paperswithcode.com/paper/extraction-of-clinical-information-from-the
Repo
Framework

Hierarchical Modeling of Multidimensional Data in Regularly Decomposed Spaces: Applications in Image Analysis


Title	Hierarchical Modeling of Multidimensional Data in Regularly Decomposed Spaces: Applications in Image Analysis
Authors	Olivier Guye
Abstract	This last document is showing the gradual introduction of hierarchical modeling techniques in image analysis. The first chapter is dealing with the first works carried out in the field of industrial applications of pattern recognition. The second chapter is focusing on the usage of these techniques in satellite imagery and on the development of a satellite data archiving system in the aim of using it in digital geography. The third chapter is about face recognition based on planar image analysis and about the recognition of partially hidden patterns. The present publication is ending with the description of a future system of self-descriptive coding of still or moving pictures in relation with the current video coding standards. As in the previous documents, it will be found in annex algorithms targeted on image analysis according two complementary approaches: - boundary-based approach for the industrial applications of artificial vision; - region-based approach for satellite image analysis.
Tasks	Face Recognition
Published	2016-05-04
URL	http://arxiv.org/abs/1605.01242v1
PDF	http://arxiv.org/pdf/1605.01242v1.pdf
PWC	https://paperswithcode.com/paper/hierarchical-modeling-of-multidimensional
Repo
Framework

Keyboard Based Control of Four Dimensional Rotations


Title	Keyboard Based Control of Four Dimensional Rotations
Authors	Akira Kageyama
Abstract	Aiming at applications to the scientific visualization of three dimensional simulations with time evolution, a keyboard based control method to specify rotations in four dimensions is proposed. It is known that four dimensional rotations are generally so-called double rotations, and a double rotation is a combination of simultaneously applied two simple rotations. The proposed method can specify both the simple and double rotations by single key typings of the keyboard. The method is tested in visualizations of a regular pentachoron in four dimensional space by a hyperplane slicing.
Tasks
Published	2016-04-06
URL	http://arxiv.org/abs/1604.02013v1
PDF	http://arxiv.org/pdf/1604.02013v1.pdf
PWC	https://paperswithcode.com/paper/keyboard-based-control-of-four-dimensional
Repo
Framework

Permuted and Augmented Stick-Breaking Bayesian Multinomial Regression


Title	Permuted and Augmented Stick-Breaking Bayesian Multinomial Regression
Authors	Quan Zhang, Mingyuan Zhou
Abstract	To model categorical response variables given their covariates, we propose a permuted and augmented stick-breaking (paSB) construction that one-to-one maps the observed categories to randomly permuted latent sticks. This new construction transforms multinomial regression into regression analysis of stick-specific binary random variables that are mutually independent given their covariate-dependent stick success probabilities, which are parameterized by the regression coefficients of their corresponding categories. The paSB construction allows transforming an arbitrary cross-entropy-loss binary classifier into a Bayesian multinomial one. Specifically, we parameterize the negative logarithms of the stick failure probabilities with a family of covariate-dependent softplus functions to construct nonparametric Bayesian multinomial softplus regression, and transform Bayesian support vector machine (SVM) into Bayesian multinomial SVM. These Bayesian multinomial regression models are not only capable of providing probability estimates, quantifying uncertainty, increasing robustness, and producing nonlinear classification decision boundaries, but also amenable to posterior simulation. Example results demonstrate their attractive properties and performance.
Tasks
Published	2016-12-30
URL	http://arxiv.org/abs/1612.09413v3
PDF	http://arxiv.org/pdf/1612.09413v3.pdf
PWC	https://paperswithcode.com/paper/permuted-and-augmented-stick-breaking
Repo
Framework

Behavior-Based Machine-Learning: A Hybrid Approach for Predicting Human Decision Making


Title	Behavior-Based Machine-Learning: A Hybrid Approach for Predicting Human Decision Making
Authors	Gali Noti, Effi Levi, Yoav Kolumbus, Amit Daniely
Abstract	A large body of work in behavioral fields attempts to develop models that describe the way people, as opposed to rational agents, make decisions. A recent Choice Prediction Competition (2015) challenged researchers to suggest a model that captures 14 classic choice biases and can predict human decisions under risk and ambiguity. The competition focused on simple decision problems, in which human subjects were asked to repeatedly choose between two gamble options. In this paper we present our approach for predicting human decision behavior: we suggest to use machine learning algorithms with features that are based on well-established behavioral theories. The basic idea is that these psychological features are essential for the representation of the data and are important for the success of the learning process. We implement a vanilla model in which we train SVM models using behavioral features that rely on the psychological properties underlying the competition baseline model. We show that this basic model captures the 14 choice biases and outperforms all the other learning-based models in the competition. The preliminary results suggest that such hybrid models can significantly improve the prediction of human decision making, and are a promising direction for future research.
Tasks	Decision Making
Published	2016-11-30
URL	http://arxiv.org/abs/1611.10228v1
PDF	http://arxiv.org/pdf/1611.10228v1.pdf
PWC	https://paperswithcode.com/paper/behavior-based-machine-learning-a-hybrid
Repo
Framework

pH Prediction by Artificial Neural Networks for the Drinking Water of the Distribution System of Hyderabad City


Title	pH Prediction by Artificial Neural Networks for the Drinking Water of the Distribution System of Hyderabad City
Authors	Niaz Ahmed Memon, Mukhtiar Ali Unar, Abdul Khalique Ansari
Abstract	In this research, feedforward ANN (Artificial Neural Network) model is developed and validated for predicting the pH at 10 different locations of the distribution system of drinking water of Hyderabad city. The developed model is MLP (Multilayer Perceptron) with back propagation algorithm.The data for the training and testing of the model are collected through an experimental analysis on weekly basis in a routine examination for maintaining the quality of drinking water in the city. 17 parameters are taken into consideration including pH. These all parameters are taken as input variables for the model and then pH is predicted for 03 phases;raw water of river Indus,treated water in the treatment plants and then treated water in the distribution system of drinking water. The training and testing results of this model reveal that MLP neural networks are exceedingly extrapolative for predicting the pH of river water, untreated and treated water at all locations of the distribution system of drinking water of Hyderabad city. The optimum input and output weights are generated with minimum MSE (Mean Square Error) < 5%.Experimental, predicted and tested values of pH are plotted and the effectiveness of the model is determined by calculating the coefficient of correlation (R2=0.999) of trained and tested results.
Tasks
Published	2016-04-02
URL	http://arxiv.org/abs/1604.00552v1
PDF	http://arxiv.org/pdf/1604.00552v1.pdf
PWC	https://paperswithcode.com/paper/ph-prediction-by-artificial-neural-networks
Repo
Framework