Paper Group ANR 363
Can mobile usage predict illiteracy in a developing country?. Decoupled Asynchronous Proximal Stochastic Gradient Descent with Variance Reduction. Learning Locomotion Skills Using DeepRL: Does the Choice of Action Space Matter?. Gaussian variational approximation with sparse precision matrices. Gated Siamese Convolutional Neural Network Architectur …
Can mobile usage predict illiteracy in a developing country?
| Title | Can mobile usage predict illiteracy in a developing country? |
| Authors | Pål Sundsøy |
| Abstract | The present study provides the first evidence that illiteracy can be reliably predicted from standard mobile phone logs. By deriving a broad set of mobile phone indicators reflecting users financial, social and mobility patterns we show how supervised machine learning can be used to predict individual illiteracy in an Asian developing country, externally validated against a large-scale survey. On average the model performs 10 times better than random guessing with a 70% accuracy. Further we show how individual illiteracy can be aggregated and mapped geographically at cell tower resolution. Geographical mapping of illiteracy is crucial to know where the illiterate people are, and where to put in resources. In underdeveloped countries such mappings are often based on out-dated household surveys with low spatial and temporal resolution. One in five people worldwide struggle with illiteracy, and it is estimated that illiteracy costs the global economy more than 1 trillion dollars each year. These results potentially enable costeffective, questionnaire-free investigation of illiteracy-related questions on an unprecedented scale |
| Tasks | |
| Published | 2016-07-05 |
| URL | http://arxiv.org/abs/1607.01337v1 |
| http://arxiv.org/pdf/1607.01337v1.pdf | |
| PWC | https://paperswithcode.com/paper/can-mobile-usage-predict-illiteracy-in-a |
| Repo | |
| Framework | |
Decoupled Asynchronous Proximal Stochastic Gradient Descent with Variance Reduction
| Title | Decoupled Asynchronous Proximal Stochastic Gradient Descent with Variance Reduction |
| Authors | Zhouyuan Huo, Bin Gu, Heng Huang |
| Abstract | In the era of big data, optimizing large scale machine learning problems becomes a challenging task and draws significant attention. Asynchronous optimization algorithms come out as a promising solution. Recently, decoupled asynchronous proximal stochastic gradient descent (DAP-SGD) is proposed to minimize a composite function. It is claimed to be able to off-loads the computation bottleneck from server to workers by allowing workers to evaluate the proximal operators, therefore, server just need to do element-wise operations. However, it still suffers from slow convergence rate because of the variance of stochastic gradient is nonzero. In this paper, we propose a faster method, decoupled asynchronous proximal stochastic variance reduced gradient descent method (DAP-SVRG). We prove that our method has linear convergence for strongly convex problem. Large-scale experiments are also conducted in this paper, and results demonstrate our theoretical analysis. |
| Tasks | |
| Published | 2016-09-22 |
| URL | http://arxiv.org/abs/1609.06804v2 |
| http://arxiv.org/pdf/1609.06804v2.pdf | |
| PWC | https://paperswithcode.com/paper/decoupled-asynchronous-proximal-stochastic |
| Repo | |
| Framework | |
Learning Locomotion Skills Using DeepRL: Does the Choice of Action Space Matter?
| Title | Learning Locomotion Skills Using DeepRL: Does the Choice of Action Space Matter? |
| Authors | Xue Bin Peng, Michiel van de Panne |
| Abstract | The use of deep reinforcement learning allows for high-dimensional state descriptors, but little is known about how the choice of action representation impacts the learning difficulty and the resulting performance. We compare the impact of four different action parameterizations (torques, muscle-activations, target joint angles, and target joint-angle velocities) in terms of learning time, policy robustness, motion quality, and policy query rates. Our results are evaluated on a gait-cycle imitation task for multiple planar articulated figures and multiple gaits. We demonstrate that the local feedback provided by higher-level action parameterizations can significantly impact the learning, robustness, and quality of the resulting policies. |
| Tasks | |
| Published | 2016-11-03 |
| URL | http://arxiv.org/abs/1611.01055v1 |
| http://arxiv.org/pdf/1611.01055v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-locomotion-skills-using-deeprl-does |
| Repo | |
| Framework | |
Gaussian variational approximation with sparse precision matrices
| Title | Gaussian variational approximation with sparse precision matrices |
| Authors | Linda S. L. Tan, David J. Nott |
| Abstract | We consider the problem of learning a Gaussian variational approximation to the posterior distribution for a high-dimensional parameter, where we impose sparsity in the precision matrix to reflect appropriate conditional independence structure in the model. Incorporating sparsity in the precision matrix allows the Gaussian variational distribution to be both flexible and parsimonious, and the sparsity is achieved through parameterization in terms of the Cholesky factor. Efficient stochastic gradient methods which make appropriate use of gradient information for the target distribution are developed for the optimization. We consider alternative estimators of the stochastic gradients which have lower variation and are more stable. Our approach is illustrated using generalized linear mixed models and state space models for time series. |
| Tasks | Time Series |
| Published | 2016-05-18 |
| URL | http://arxiv.org/abs/1605.05622v3 |
| http://arxiv.org/pdf/1605.05622v3.pdf | |
| PWC | https://paperswithcode.com/paper/gaussian-variational-approximation-with |
| Repo | |
| Framework | |
Gated Siamese Convolutional Neural Network Architecture for Human Re-Identification
| Title | Gated Siamese Convolutional Neural Network Architecture for Human Re-Identification |
| Authors | Rahul Rama Varior, Mrinal Haloi, Gang Wang |
| Abstract | Matching pedestrians across multiple camera views, known as human re-identification, is a challenging research problem that has numerous applications in visual surveillance. With the resurgence of Convolutional Neural Networks (CNNs), several end-to-end deep Siamese CNN architectures have been proposed for human re-identification with the objective of projecting the images of similar pairs (i.e. same identity) to be closer to each other and those of dissimilar pairs to be distant from each other. However, current networks extract fixed representations for each image regardless of other images which are paired with it and the comparison with other images is done only at the final level. In this setting, the network is at risk of failing to extract finer local patterns that may be essential to distinguish positive pairs from hard negative pairs. In this paper, we propose a gating function to selectively emphasize such fine common local patterns by comparing the mid-level features across pairs of images. This produces flexible representations for the same image according to the images they are paired with. We conduct experiments on the CUHK03, Market-1501 and VIPeR datasets and demonstrate improved performance compared to a baseline Siamese CNN architecture. |
| Tasks | Person Re-Identification |
| Published | 2016-07-28 |
| URL | http://arxiv.org/abs/1607.08378v2 |
| http://arxiv.org/pdf/1607.08378v2.pdf | |
| PWC | https://paperswithcode.com/paper/gated-siamese-convolutional-neural-network |
| Repo | |
| Framework | |
Learning Transferable Policies for Monocular Reactive MAV Control
| Title | Learning Transferable Policies for Monocular Reactive MAV Control |
| Authors | Shreyansh Daftry, J. Andrew Bagnell, Martial Hebert |
| Abstract | The ability to transfer knowledge gained in previous tasks into new contexts is one of the most important mechanisms of human learning. Despite this, adapting autonomous behavior to be reused in partially similar settings is still an open problem in current robotics research. In this paper, we take a small step in this direction and propose a generic framework for learning transferable motion policies. Our goal is to solve a learning problem in a target domain by utilizing the training data in a different but related source domain. We present this in the context of an autonomous MAV flight using monocular reactive control, and demonstrate the efficacy of our proposed approach through extensive real-world flight experiments in outdoor cluttered environments. |
| Tasks | |
| Published | 2016-08-01 |
| URL | http://arxiv.org/abs/1608.00627v1 |
| http://arxiv.org/pdf/1608.00627v1.pdf | |
| PWC | https://paperswithcode.com/paper/learning-transferable-policies-for-monocular |
| Repo | |
| Framework | |
Hard-Aware Deeply Cascaded Embedding
| Title | Hard-Aware Deeply Cascaded Embedding |
| Authors | Yuhui Yuan, Kuiyuan Yang, Chao Zhang |
| Abstract | Riding on the waves of deep neural networks, deep metric learning has also achieved promising results in various tasks using triplet network or Siamese network. Though the basic goal of making images from the same category closer than the ones from different categories is intuitive, it is hard to directly optimize due to the quadratic or cubic sample size. To solve the problem, hard example mining which only focuses on a subset of samples that are considered hard is widely used. However, hard is defined relative to a model, where complex models treat most samples as easy ones and vice versa for simple models, and both are not good for training. Samples are also with different hard levels, it is hard to define a model with the just right complexity and choose hard examples adequately. This motivates us to ensemble a set of models with different complexities in cascaded manner and mine hard examples adaptively, a sample is judged by a series of models with increasing complexities and only updates models that consider the sample as a hard case. We evaluate our method on CARS196, CUB-200-2011, Stanford Online Products, VehicleID and DeepFashion datasets. Our method outperforms state-of-the-art methods by a large margin. |
| Tasks | Metric Learning |
| Published | 2016-11-17 |
| URL | http://arxiv.org/abs/1611.05720v2 |
| http://arxiv.org/pdf/1611.05720v2.pdf | |
| PWC | https://paperswithcode.com/paper/hard-aware-deeply-cascaded-embedding |
| Repo | |
| Framework | |
Open-Ended Visual Question-Answering
| Title | Open-Ended Visual Question-Answering |
| Authors | Issey Masuda, Santiago Pascual de la Puente, Xavier Giro-i-Nieto |
| Abstract | This thesis report studies methods to solve Visual Question-Answering (VQA) tasks with a Deep Learning framework. As a preliminary step, we explore Long Short-Term Memory (LSTM) networks used in Natural Language Processing (NLP) to tackle Question-Answering (text based). We then modify the previous model to accept an image as an input in addition to the question. For this purpose, we explore the VGG-16 and K-CNN convolutional neural networks to extract visual features from the image. These are merged with the word embedding or with a sentence embedding of the question to predict the answer. This work was successfully submitted to the Visual Question Answering Challenge 2016, where it achieved a 53,62% of accuracy in the test dataset. The developed software has followed the best programming practices and Python code style, providing a consistent baseline in Keras for different configurations. |
| Tasks | Question Answering, Sentence Embedding, Visual Question Answering |
| Published | 2016-10-09 |
| URL | http://arxiv.org/abs/1610.02692v1 |
| http://arxiv.org/pdf/1610.02692v1.pdf | |
| PWC | https://paperswithcode.com/paper/open-ended-visual-question-answering |
| Repo | |
| Framework | |
Real-time Halfway Domain Reconstruction of Motion and Geometry
| Title | Real-time Halfway Domain Reconstruction of Motion and Geometry |
| Authors | Lucas Thies, Michael Zollhöfer, Christian Richardt, Christian Theobalt, Günther Greiner |
| Abstract | We present a novel approach for real-time joint reconstruction of 3D scene motion and geometry from binocular stereo videos. Our approach is based on a novel variational halfway-domain scene flow formulation, which allows us to obtain highly accurate spatiotemporal reconstructions of shape and motion. We solve the underlying optimization problem at real-time frame rates using a novel data-parallel robust non-linear optimization strategy. Fast convergence and large displacement flows are achieved by employing a novel hierarchy that stores delta flows between hierarchy levels. High performance is obtained by the introduction of a coarser warp grid that decouples the number of unknowns from the input resolution of the images. We demonstrate our approach in a live setup that is based on two commodity webcams, as well as on publicly available video data. Our extensive experiments and evaluations show that our approach produces high-quality dense reconstructions of 3D geometry and scene flow at real-time frame rates, and compares favorably to the state of the art. |
| Tasks | |
| Published | 2016-10-23 |
| URL | http://arxiv.org/abs/1610.07159v1 |
| http://arxiv.org/pdf/1610.07159v1.pdf | |
| PWC | https://paperswithcode.com/paper/real-time-halfway-domain-reconstruction-of |
| Repo | |
| Framework | |
Doubly Accelerated Methods for Faster CCA and Generalized Eigendecomposition
| Title | Doubly Accelerated Methods for Faster CCA and Generalized Eigendecomposition |
| Authors | Zeyuan Allen-Zhu, Yuanzhi Li |
| Abstract | We study $k$-GenEV, the problem of finding the top $k$ generalized eigenvectors, and $k$-CCA, the problem of finding the top $k$ vectors in canonical-correlation analysis. We propose algorithms $\mathtt{LazyEV}$ and $\mathtt{LazyCCA}$ to solve the two problems with running times linearly dependent on the input size and on $k$. Furthermore, our algorithms are DOUBLY-ACCELERATED: our running times depend only on the square root of the matrix condition number, and on the square root of the eigengap. This is the first such result for both $k$-GenEV or $k$-CCA. We also provide the first gap-free results, which provide running times that depend on $1/\sqrt{\varepsilon}$ rather than the eigengap. |
| Tasks | |
| Published | 2016-07-20 |
| URL | http://arxiv.org/abs/1607.06017v2 |
| http://arxiv.org/pdf/1607.06017v2.pdf | |
| PWC | https://paperswithcode.com/paper/doubly-accelerated-methods-for-faster-cca-and |
| Repo | |
| Framework | |
Automatic Image De-fencing System
| Title | Automatic Image De-fencing System |
| Authors | Krishna Kanth Nakka |
| Abstract | Tourists and Wild-life photographers are often hindered in capturing their cherished images or videos by a fence that limits accessibility to the scene of interest. The situation has been exacerbated by growing concerns of security at public places and a need exists to provide a tool that can be used for post-processing such fenced videos to produce a de-fenced image. There are several challenges in this problem, we identify them as Robust detection of fence/occlusions and Estimating pixel motion of background scenes and Filling in the fence/occlusions by utilizing information in multiple frames of the input video. In this work, we aim to build an automatic post-processing tool that can efficiently rid the input video of occlusion artifacts like fences. Our work is distinguished by two major contributions. The first is the introduction of learning based technique to detect the fences patterns with complicated backgrounds. The second is the formulation of objective function and further minimization through loopy belief propagation to fill-in the fence pixels. We observe that grids of Histogram of oriented gradients descriptor using Support vector machines based classifier significantly outperforms detection accuracy of texels in a lattice. We present results of experiments using several real-world videos to demonstrate the effectiveness of the proposed fence detection and de-fencing algorithm. |
| Tasks | |
| Published | 2016-10-21 |
| URL | http://arxiv.org/abs/1610.06924v1 |
| http://arxiv.org/pdf/1610.06924v1.pdf | |
| PWC | https://paperswithcode.com/paper/automatic-image-de-fencing-system |
| Repo | |
| Framework | |
Quantum cognition beyond Hilbert space II: Applications
| Title | Quantum cognition beyond Hilbert space II: Applications |
| Authors | Diederik Aerts, Lyneth Beltran, Massimiliano Sassoli de Bianchi, Sandro Sozzo, Tomas Veloz |
| Abstract | The research on human cognition has recently benefited from the use of the mathematical formalism of quantum theory in Hilbert space. However, cognitive situations exist which indicate that the Hilbert space structure, and the associated Born rule, would be insufficient to provide a satisfactory modeling of the collected data, so that one needs to go beyond Hilbert space. In Part I of this paper we follow this direction and present a general tension-reduction (GTR) model, in the ambit of an operational and realistic framework for human cognition. In this Part II we apply this non-Hilbertian quantum-like model to faithfully reproduce the probabilities of the ‘Clinton/Gore’ and ‘Rose/Jackson’ experiments on question order effects. We also explain why the GTR-model is needed if one wants to deal, in a fully consistent way, with response replicability and unpacking effects. |
| Tasks | |
| Published | 2016-04-27 |
| URL | http://arxiv.org/abs/1604.08270v1 |
| http://arxiv.org/pdf/1604.08270v1.pdf | |
| PWC | https://paperswithcode.com/paper/quantum-cognition-beyond-hilbert-space-ii |
| Repo | |
| Framework | |
Kernel Mean Embedding of Distributions: A Review and Beyond
| Title | Kernel Mean Embedding of Distributions: A Review and Beyond |
| Authors | Krikamol Muandet, Kenji Fukumizu, Bharath Sriperumbudur, Bernhard Schölkopf |
| Abstract | A Hilbert space embedding of a distribution—in short, a kernel mean embedding—has recently emerged as a powerful tool for machine learning and inference. The basic idea behind this framework is to map distributions into a reproducing kernel Hilbert space (RKHS) in which the whole arsenal of kernel methods can be extended to probability measures. It can be viewed as a generalization of the original “feature map” common to support vector machines (SVMs) and other kernel methods. While initially closely associated with the latter, it has meanwhile found application in fields ranging from kernel machines and probabilistic modeling to statistical inference, causal discovery, and deep learning. The goal of this survey is to give a comprehensive review of existing work and recent advances in this research area, and to discuss the most challenging issues and open problems that could lead to new research directions. The survey begins with a brief introduction to the RKHS and positive definite kernels which forms the backbone of this survey, followed by a thorough discussion of the Hilbert space embedding of marginal distributions, theoretical guarantees, and a review of its applications. The embedding of distributions enables us to apply RKHS methods to probability measures which prompts a wide range of applications such as kernel two-sample testing, independent testing, and learning on distributional data. Next, we discuss the Hilbert space embedding for conditional distributions, give theoretical insights, and review some applications. The conditional mean embedding enables us to perform sum, product, and Bayes’ rules—which are ubiquitous in graphical model, probabilistic inference, and reinforcement learning—in a non-parametric way. We then discuss relationships between this framework and other related areas. Lastly, we give some suggestions on future research directions. |
| Tasks | Causal Discovery |
| Published | 2016-05-31 |
| URL | http://arxiv.org/abs/1605.09522v3 |
| http://arxiv.org/pdf/1605.09522v3.pdf | |
| PWC | https://paperswithcode.com/paper/kernel-mean-embedding-of-distributions-a |
| Repo | |
| Framework | |
Application of the Second-Order Statistics for Estimation of the Pure Spectra of Individual Components from the Visible Hyperspectral Images of Their Mixture
| Title | Application of the Second-Order Statistics for Estimation of the Pure Spectra of Individual Components from the Visible Hyperspectral Images of Their Mixture |
| Authors | Sung-Ho Jong, Yong-U Ri, Kye-Ryong Sin |
| Abstract | The second-order statistics (SOS) can be applied in estimation of the pure spectra of chemical components from the spectrum of their mixture, when SOS seems to be good at estimation of spectral patterns, but their peak directions are opposite in some cases. In this paper, one method for judgment of the peak direction of the pure spectra was proposed, where the base line of the pure spectra was drawn by using their histograms and the peak directions were chosen so as to make all of the pure spectra located upwards over the base line. Results of the SOS analysis on the visible hyperspectral images of the mixture composed of two or three chemical components showed that the present method offered the reasonable shape and direction of the pure spectra of its components. |
| Tasks | |
| Published | 2016-04-12 |
| URL | http://arxiv.org/abs/1604.03193v1 |
| http://arxiv.org/pdf/1604.03193v1.pdf | |
| PWC | https://paperswithcode.com/paper/application-of-the-second-order-statistics |
| Repo | |
| Framework | |
On the Expressive Power of Deep Neural Networks
| Title | On the Expressive Power of Deep Neural Networks |
| Authors | Maithra Raghu, Ben Poole, Jon Kleinberg, Surya Ganguli, Jascha Sohl-Dickstein |
| Abstract | We propose a new approach to the problem of neural network expressivity, which seeks to characterize how structural properties of a neural network family affect the functions it is able to compute. Our approach is based on an interrelated set of measures of expressivity, unified by the novel notion of trajectory length, which measures how the output of a network changes as the input sweeps along a one-dimensional path. Our findings can be summarized as follows: (1) The complexity of the computed function grows exponentially with depth. (2) All weights are not equal: trained networks are more sensitive to their lower (initial) layer weights. (3) Regularizing on trajectory length (trajectory regularization) is a simpler alternative to batch normalization, with the same performance. |
| Tasks | |
| Published | 2016-06-16 |
| URL | http://arxiv.org/abs/1606.05336v6 |
| http://arxiv.org/pdf/1606.05336v6.pdf | |
| PWC | https://paperswithcode.com/paper/on-the-expressive-power-of-deep-neural |
| Repo | |
| Framework | |
