January 27, 2020

2976 words 14 mins read

Paper Group ANR 1249

Distributed Nesterov gradient methods over arbitrary graphs. Identifying nearby sources of ultra-high-energy cosmic rays with deep learning. First Order Ambisonics Domain Spatial Augmentation for DNN-based Direction of Arrival Estimation. Texture Bias Of CNNs Limits Few-Shot Classification Performance. Scalable Change Retrieval Using Deep 3D Neural …

Distributed Nesterov gradient methods over arbitrary graphs


Title	Distributed Nesterov gradient methods over arbitrary graphs
Authors	Ran Xin, Dusan Jakovetic, Usman A. Khan
Abstract	In this letter, we introduce a distributed Nesterov method, termed as $\mathcal{ABN}$, that does not require doubly-stochastic weight matrices. Instead, the implementation is based on a simultaneous application of both row- and column-stochastic weights that makes this method applicable to arbitrary (strongly-connected) graphs. Since constructing column-stochastic weights needs additional information (the number of outgoing neighbors at each agent), not available in certain communication protocols, we derive a variation, termed as FROZEN, that only requires row-stochastic weights but at the expense of additional iterations for eigenvector learning. We numerically study these algorithms for various objective functions and network parameters and show that the proposed distributed Nesterov methods achieve acceleration compared to the current state-of-the-art methods for distributed optimization.
Tasks	Distributed Optimization
Published	2019-01-21
URL	http://arxiv.org/abs/1901.06995v1
PDF	http://arxiv.org/pdf/1901.06995v1.pdf
PWC	https://paperswithcode.com/paper/distributed-nesterov-gradient-methods-over
Repo
Framework

Identifying nearby sources of ultra-high-energy cosmic rays with deep learning


Title	Identifying nearby sources of ultra-high-energy cosmic rays with deep learning
Authors	Oleg Kalashev, Maxim Pshirkov, Mikhail Zotov
Abstract	We present a method to analyse arrival directions of ultra-high-energy cosmic rays (UHECRs) using a classifier defined by a deep convolutional neural network trained on a HEALPix grid. To illustrate the efficacy of the method, we employ it to estimate prospects of detecting a large-scale anisotropy of UHECRs induced by a nearby source with an (orbital) detector having a uniform exposure of the celestial sphere and compare the results with our earlier calculations based on the angular power spectrum. A minimal model for extragalactic cosmic rays and neutrinos by Kachelrie{\ss}, Kalashev, Ostapchenko and Semikoz (2017) is assumed for definiteness and nearby active galactic nuclei Centaurus A, M82, NGC253, M87 and Fornax A are considered as possible sources of UHECRs. We demonstrate that the proposed method drastically improves sensitivity of an experiment by decreasing the minimal required amount of detected UHECRs or the minimal detectable fraction of from-source events several times compared to the approach based on the angular power spectrum. The method can be readily applied to the analysis of data of the Telescope Array, the Pierre Auger Observatory and other cosmic ray experiments.
Tasks
Published	2019-12-02
URL	https://arxiv.org/abs/1912.00625v1
PDF	https://arxiv.org/pdf/1912.00625v1.pdf
PWC	https://paperswithcode.com/paper/identifying-nearby-sources-of-ultra-high
Repo
Framework

First Order Ambisonics Domain Spatial Augmentation for DNN-based Direction of Arrival Estimation


Title	First Order Ambisonics Domain Spatial Augmentation for DNN-based Direction of Arrival Estimation
Authors	Luca Mazzon, Yuma Koizumi, Masahiro Yasuda, Noboru Harada
Abstract	In this paper, we propose a novel data augmentation method for training neural networks for Direction of Arrival (DOA) estimation. This method focuses on expanding the representation of the DOA subspace of a dataset. Given some input data, it applies a transformation to it in order to change its DOA information and simulate new potentially unseen one. Such transformation, in general, is a combination of a rotation and a reflection. It is possible to apply such transformation due to a well-known property of First Order Ambisonics (FOA). The same transformation is applied also to the labels, in order to maintain consistency between input data and target labels. Three methods with different level of generality are proposed for applying this augmentation principle. Experiments are conducted on two different DOA networks. Results of both experiments demonstrate the effectiveness of the novel augmentation strategy by improving the DOA error by around 40%.
Tasks	Data Augmentation, Direction of Arrival Estimation
Published	2019-10-10
URL	https://arxiv.org/abs/1910.04388v1
PDF	https://arxiv.org/pdf/1910.04388v1.pdf
PWC	https://paperswithcode.com/paper/first-order-ambisonics-domain-spatial
Repo
Framework

Texture Bias Of CNNs Limits Few-Shot Classification Performance


Title	Texture Bias Of CNNs Limits Few-Shot Classification Performance
Authors	Sam Ringer, Will Williams, Tom Ash, Remi Francis, David MacLeod
Abstract	Accurate image classification given small amounts of labelled data (few-shot classification) remains an open problem in computer vision. In this work we examine how the known texture bias of Convolutional Neural Networks (CNNs) affects few-shot classification performance. Although texture bias can help in standard image classification, in this work we show it significantly harms few-shot classification performance. After correcting this bias we demonstrate state-of-the-art performance on the competitive miniImageNet task using a method far simpler than the current best performing few-shot learning approaches.
Tasks	Few-Shot Learning, Image Classification
Published	2019-10-18
URL	https://arxiv.org/abs/1910.08519v1
PDF	https://arxiv.org/pdf/1910.08519v1.pdf
PWC	https://paperswithcode.com/paper/texture-bias-of-cnns-limits-few-shot
Repo
Framework

Scalable Change Retrieval Using Deep 3D Neural Codes


Title	Scalable Change Retrieval Using Deep 3D Neural Codes
Authors	Kojima Yusuke, Tanaka Kanji, Yang Naiming, Hirota Yuji
Abstract	We present a novel scalable framework for image change detection (ICD) from an on-board 3D imagery system. We argue that existing ICD systems are constrained by the time required to align a given query image with individual reference image coordinates. We utilize an invariant coordinate system (ICS) to replace the time-consuming image alignment with an offline pre-processing procedure. Our key contribution is an extension of the traditional image comparison-based ICD tasks to setups of the image retrieval (IR) task. We replace each component of the 3D ICD system, i.e., (1) image modeling, (2) image alignment, and (3) image differencing, with significantly efficient variants from the bag-of-words (BoW) IR paradigm. Further, we train a deep 3D feature extractor in an unsupervised manner using an unsupervised Siamese network and automatically collected training data. We conducted experiments on a challenging cross-season ICD task using a publicly available dataset and thereby validate the efficacy of the proposed approach.
Tasks	Image Retrieval
Published	2019-04-07
URL	http://arxiv.org/abs/1904.03552v1
PDF	http://arxiv.org/pdf/1904.03552v1.pdf
PWC	https://paperswithcode.com/paper/scalable-change-retrieval-using-deep-3d
Repo
Framework

CBHE: Corner-based Building Height Estimation for Complex Street Scene Images


Title	CBHE: Corner-based Building Height Estimation for Complex Street Scene Images
Authors	Yunxiang Zhao, Jianzhong Qi, Rui Zhang
Abstract	Building height estimation is important in many applications such as 3D city reconstruction, urban planning, and navigation. Recently, a new building height estimation method using street scene images and 2D maps was proposed. This method is more scalable than traditional methods that use high-resolution optical data, LiDAR data, or RADAR data which are expensive to obtain. The method needs to detect building rooflines and then compute building height via the pinhole camera model. We observe that this method has limitations in handling complex street scene images in which buildings overlap with each other and the rooflines are difficult to locate. We propose CBHE, a building height estimation algorithm considering both building corners and rooflines. CBHE first obtains building corner and roofline candidates in street scene images based on building footprints from 2D maps and the camera parameters. Then, we use a deep neural network named BuildingNet to classify and filter corner and roofline candidates. Based on the valid corners and rooflines from BuildingNet, CBHE computes building height via the pinhole camera model. Experimental results show that the proposed BuildingNet yields a higher accuracy on building corner and roofline candidate filtering compared with the state-of-the-art open set classifiers. Meanwhile, CBHE outperforms the baseline algorithm by over 10% in building height estimation accuracy.
Tasks
Published	2019-04-25
URL	http://arxiv.org/abs/1904.11128v1
PDF	http://arxiv.org/pdf/1904.11128v1.pdf
PWC	https://paperswithcode.com/paper/cbhe-corner-based-building-height-estimation
Repo
Framework

Ranking with concept drift. Weighted tournament solutions for evolving preferences


Title	Ranking with concept drift. Weighted tournament solutions for evolving preferences
Authors	Ekhine Irurozki, Jesus Lobo, Aritz Perez, Javier Del Ser
Abstract	This work deals with the problem of rank aggregation over non-stationary ranking streams. The rankings can be interpreted as noisy realizations of an unknown probability distribution that changes over time. Our goal is to learn, in an online manner, the current ground truth distribution of rankings, which essentially translates to a rank aggregation problem. First, we generalize the family of weighted voting rules to situations in which some rankings are more \textit{reliable} than others and show that this generalization can solve the problem of rank aggregation over non-stationary data streams. Next, we propose unbalanced Borda for non-stationary ranking streams, and we give bounds on the minimum number of samples required to output the ground truth with high probability. Finally, we show an application of the adaptation to recommender systems.
Tasks	Recommendation Systems
Published	2019-10-19
URL	https://arxiv.org/abs/1910.08795v2
PDF	https://arxiv.org/pdf/1910.08795v2.pdf
PWC	https://paperswithcode.com/paper/online-ranking-with-concept-drifts-in
Repo
Framework

Privacy-Preserving Bandits


Title	Privacy-Preserving Bandits
Authors	Mohammad Malekzadeh, Dimitrios Athanasakis, Hamed Haddadi, Benjamin Livshits
Abstract	Contextual bandit algorithms~(CBAs) often rely on personal data to provide recommendations. Centralized CBA agents utilize potentially sensitive data from recent interactions to provide personalization to end-users. Keeping the sensitive data locally, by running a local agent on the user’s device, protects the user’s privacy, however, the agent requires longer to produce useful recommendations, as it does not leverage feedback from other users. This paper proposes a technique we call Privacy-Preserving Bandits (P2B); a system that updates local agents by collecting feedback from other local agents in a differentially-private manner. Comparisons of our proposed approach with a non-private, as well as a fully-private (local) system, show competitive performance on both synthetic benchmarks and real-world data. Specifically, we observed only a decrease of 2.6% and 3.6% in multi-label classification accuracy, and a CTR increase of 0.0025 in online advertising for a privacy budget $\epsilon \approx 0.693$. These results suggest P2B is an effective approach to challenges arising in on-device privacy-preserving personalization.
Tasks	Multi-Label Classification
Published	2019-09-10
URL	https://arxiv.org/abs/1909.04421v4
PDF	https://arxiv.org/pdf/1909.04421v4.pdf
PWC	https://paperswithcode.com/paper/privacy-preserving-bandits
Repo
Framework

A Closer Look at Mobile App Usage as a Persistent Biometric: A Small Case Study


Title	A Closer Look at Mobile App Usage as a Persistent Biometric: A Small Case Study
Authors	Md A. Noor, G. Kaptan, V. Cherukupally, P. Gera, T. Neal
Abstract	In this paper, we explore mobile app use as a behavioral biometric identifier. While several efforts have also taken on this challenge, many have alluded to the inconsistency in human behavior, resulting in updating the biometric template frequently and periodically. Here, we represent app usage as simple images wherein each pixel value provides some information about the user’s app usage. Then, we feed use these images to train a deep learning network (convolutional neural net) to classify the user’s identity. Our contribution lies in the random order in which the images are fed to the classifier, thereby presenting novel evidence that there are some aspects of app usage that are indeed persistent. Our results yield a 96.8% $F$-score without any updates to the template data.
Tasks
Published	2019-12-25
URL	https://arxiv.org/abs/1912.11721v1
PDF	https://arxiv.org/pdf/1912.11721v1.pdf
PWC	https://paperswithcode.com/paper/a-closer-look-at-mobile-app-usage-as-a
Repo
Framework

Active Image Synthesis for Efficient Labeling


Title	Active Image Synthesis for Efficient Labeling
Authors	Jialei Chen, Yujia Xie, Kan Wang, Chuck Zhang, Mani A. Vannan, Ben Wang, Zhen Qian
Abstract	The great success achieved by deep neural networks attracts increasing attention from the manufacturing and healthcare communities. However, the limited availability of data and high costs of data collection are the major challenges for the applications in those fields. We propose in this work AISEL, an active image synthesis method for efficient labeling to improve the performance of the small-data learning tasks. Specifically, a complementary AISEL dataset is generated, with labels actively acquired via a physics-based method to incorporate underlining physical knowledge at hand. An important component of our AISEL method is the bidirectional generative invertible network (GIN), which can extract interpretable features from the training images and generate physically meaningful virtual images. Our AISEL method then efficiently samples virtual images not only further exploits the uncertain regions, but also explores the entire image space. We then discuss the interpretability of GIN both theoretically and experimentally, demonstrating clear visual improvements over the benchmarks. Finally, we demonstrate the effectiveness of our AISEL framework on aortic stenosis application, in which our method lower the labeling cost by $90%$ while achieving a $15%$ improvement in prediction accuracy.
Tasks	Image Generation
Published	2019-02-05
URL	https://arxiv.org/abs/1902.01522v3
PDF	https://arxiv.org/pdf/1902.01522v3.pdf
PWC	https://paperswithcode.com/paper/avp-physics-informed-data-generation-for
Repo
Framework

Testing Independence with the Binary Expansion Randomized Ensemble Test


Title	Testing Independence with the Binary Expansion Randomized Ensemble Test
Authors	Duyeol Lee, Kai Zhang, Michael R. Kosorok
Abstract	Recently, the binary expansion testing framework was introduced to test the independence of two continuous random variables by utilizing symmetry statistics that are complete sufficient statistics for dependence. We develop a new test by an ensemble method that uses the sum of squared symmetry statistics and distance correlation. Simulation studies suggest that this method improves the power while preserving the clear interpretation of the binary expansion testing. We extend this method to tests of independence of random vectors in arbitrary dimension. By random projections, the proposed binary expansion randomized ensemble test transforms the multivariate independence testing problem into a univariate problem. Simulation studies and data example analyses show that the proposed method provides relatively robust performance compared with existing methods.
Tasks
Published	2019-12-08
URL	https://arxiv.org/abs/1912.03662v2
PDF	https://arxiv.org/pdf/1912.03662v2.pdf
PWC	https://paperswithcode.com/paper/testing-independence-with-the-binary
Repo
Framework

Discriminative Local Sparse Representation by Robust Adaptive Dictionary Pair Learning


Title	Discriminative Local Sparse Representation by Robust Adaptive Dictionary Pair Learning
Authors	Yulin Sun, Zhao Zhang, Weiming Jiang, Zheng Zhang, Li Zhang, Shuicheng Yan, Meng Wang
Abstract	In this paper, we propose a structured Robust Adaptive Dic-tionary Pair Learning (RA-DPL) framework for the discrim-inative sparse representation learning. To achieve powerful representation ability of the available samples, the setting of RA-DPL seamlessly integrates the robust projective dictionary pair learning, locality-adaptive sparse representations and discriminative coding coefficients learning into a unified learning framework. Specifically, RA-DPL improves existing projective dictionary pair learning in four perspectives. First, it applies a sparse l2,1-norm based metric to encode the recon-struction error to deliver the robust projective dictionary pairs, and the l2,1-norm has the potential to minimize the error. Sec-ond, it imposes the robust l2,1-norm clearly on the analysis dictionary to ensure the sparse property of the coding coeffi-cients rather than using the costly l0/l1-norm. As such, the robustness of the data representation and the efficiency of the learning process are jointly considered to guarantee the effi-cacy of our RA-DPL. Third, RA-DPL conceives a structured reconstruction weight learning paradigm to preserve the local structures of the coding coefficients within each class clearly in an adaptive manner, which encourages to produce the locality preserving representations. Fourth, it also considers improving the discriminating ability of coding coefficients and dictionary by incorporating a discriminating function, which can ensure high intra-class compactness and inter-class separation in the code space. Extensive experiments show that our RA-DPL can obtain superior performance over other state-of-the-arts.
Tasks	Representation Learning
Published	2019-11-20
URL	https://arxiv.org/abs/1911.08680v1
PDF	https://arxiv.org/pdf/1911.08680v1.pdf
PWC	https://paperswithcode.com/paper/discriminative-local-sparse-representation-by
Repo
Framework

Crowdsourced PAC Learning under Classification Noise


Title	Crowdsourced PAC Learning under Classification Noise
Authors	Shelby Heinecke, Lev Reyzin
Abstract	In this paper, we analyze PAC learnability from labels produced by crowdsourcing. In our setting, unlabeled examples are drawn from a distribution and labels are crowdsourced from workers who operate under classification noise, each with their own noise parameter. We develop an end-to-end crowdsourced PAC learning algorithm that takes unlabeled data points as input and outputs a trained classifier. Our three-step algorithm incorporates majority voting, pure-exploration bandits, and noisy-PAC learning. We prove several guarantees on the number of tasks labeled by workers for PAC learning in this setting and show that our algorithm improves upon the baseline by reducing the total number of tasks given to workers. We demonstrate the robustness of our algorithm by exploring its application to additional realistic crowdsourcing settings.
Tasks
Published	2019-02-12
URL	http://arxiv.org/abs/1902.04629v1
PDF	http://arxiv.org/pdf/1902.04629v1.pdf
PWC	https://paperswithcode.com/paper/crowdsourced-pac-learning-under
Repo
Framework

TossingBot: Learning to Throw Arbitrary Objects with Residual Physics


Title	TossingBot: Learning to Throw Arbitrary Objects with Residual Physics
Authors	Andy Zeng, Shuran Song, Johnny Lee, Alberto Rodriguez, Thomas Funkhouser
Abstract	We investigate whether a robot arm can learn to pick and throw arbitrary objects into selected boxes quickly and accurately. Throwing has the potential to increase the physical reachability and picking speed of a robot arm. However, precisely throwing arbitrary objects in unstructured settings presents many challenges: from acquiring reliable pre-throw conditions (e.g. initial pose of object in manipulator) to handling varying object-centric properties (e.g. mass distribution, friction, shape) and dynamics (e.g. aerodynamics). In this work, we propose an end-to-end formulation that jointly learns to infer control parameters for grasping and throwing motion primitives from visual observations (images of arbitrary objects in a bin) through trial and error. Within this formulation, we investigate the synergies between grasping and throwing (i.e., learning grasps that enable more accurate throws) and between simulation and deep learning (i.e., using deep networks to predict residuals on top of control parameters predicted by a physics simulator). The resulting system, TossingBot, is able to grasp and throw arbitrary objects into boxes located outside its maximum reach range at 500+ mean picks per hour (600+ grasps per hour with 85% throwing accuracy); and generalizes to new objects and target locations. Videos are available at https://tossingbot.cs.princeton.edu
Tasks
Published	2019-03-27
URL	https://arxiv.org/abs/1903.11239v2
PDF	https://arxiv.org/pdf/1903.11239v2.pdf
PWC	https://paperswithcode.com/paper/tossingbot-learning-to-throw-arbitrary
Repo
Framework

Universal adversarial examples in speech command classification


Title	Universal adversarial examples in speech command classification
Authors	Jon Vadillo, Roberto Santana
Abstract	Adversarial examples are inputs intentionally perturbed with the aim of forcing a machine learning model to produce a wrong prediction, while the changes are not easily detectable by a human. Although this topic has been intensively studied in the image domain, classification tasks in the audio domain have received less attention. In this paper we address the existence of universal perturbations for speech command classification. We provide evidence that universal attacks can be generated for speech command classification tasks, which are able to generalize across different models to a significant extent. Additionally, a novel analytical framework is proposed for the evaluation of universal perturbations under different levels of universality, demonstrating that the feasibility of generating effective perturbations decreases as the universality level increases. Finally, we propose a more detailed and rigorous framework to measure the amount of distortion introduced by the perturbations, demonstrating that the methods employed by convention are not realistic in audio-based problems.
Tasks
Published	2019-11-22
URL	https://arxiv.org/abs/1911.10182v3
PDF	https://arxiv.org/pdf/1911.10182v3.pdf
PWC	https://paperswithcode.com/paper/universal-adversarial-examples-in-speech
Repo
Framework