April 3, 2020

3370 words 16 mins read

Paper Group AWR 29

Paper Group AWR 29

A Nonconvex Low-Rank Tensor Completion Model for Spatiotemporal Traffic Data Imputation. Memory Enhanced Global-Local Aggregation for Video Object Detection. Design, Validation, and Case Studies of 2D-VSR-Sim, an Optimization-friendly Simulator of 2-D Voxel-based Soft Robots. Efficient Rule Learning with Template Saturation for Knowledge Graph Comp …

A Nonconvex Low-Rank Tensor Completion Model for Spatiotemporal Traffic Data Imputation

Title A Nonconvex Low-Rank Tensor Completion Model for Spatiotemporal Traffic Data Imputation
Authors Xinyu Chen, Jinming Yang, Lijun Sun
Abstract Sparsity and missing data problems are very common in spatiotemporal traffic data collected from various sensing systems. Making accurate imputation is critical to many applications in intelligent transportation systems. In this paper, we formulate the missing data imputation problem in spatiotemporal traffic data in a low-rank tensor completion (LRTC) framework and define a novel truncated nuclear norm (TNN) on traffic tensors of location$\times$day$\times$time of day. In particular, we introduce an universal rate parameter to control the degree of truncation on all tensor modes in the proposed LRTC-TNN model, and this allows us to better characterize the hidden patterns in spatiotemporal traffic data. Based on the framework of the Alternating Direction Method of Multipliers (ADMM), we present an efficient algorithm to obtain the optimal solution for each variable. We conduct numerical experiments on four spatiotemporal traffic data sets, and our results show that the proposed LRTC-TNN model outperforms many state-of-the-art imputation models with missing rates/patterns. Moreover, the proposed model also outperforms other baseline models in extreme missing scenarios.
Tasks Imputation, Traffic Data Imputation
Published 2020-03-23
URL https://arxiv.org/abs/2003.10271v1
PDF https://arxiv.org/pdf/2003.10271v1.pdf
PWC https://paperswithcode.com/paper/a-nonconvex-low-rank-tensor-completion-model
Repo https://github.com/xinychen/transdim
Framework tf

Memory Enhanced Global-Local Aggregation for Video Object Detection

Title Memory Enhanced Global-Local Aggregation for Video Object Detection
Authors Yihong Chen, Yue Cao, Han Hu, Liwei Wang
Abstract How do humans recognize an object in a piece of video? Due to the deteriorated quality of single frame, it may be hard for people to identify an occluded object in this frame by just utilizing information within one image. We argue that there are two important cues for humans to recognize objects in videos: the global semantic information and the local localization information. Recently, plenty of methods adopt the self-attention mechanisms to enhance the features in key frame with either global semantic information or local localization information. In this paper we introduce memory enhanced global-local aggregation (MEGA) network, which is among the first trials that takes full consideration of both global and local information. Furthermore, empowered by a novel and carefully-designed Long Range Memory (LRM) module, our proposed MEGA could enable the key frame to get access to much more content than any previous methods. Enhanced by these two sources of information, our method achieves state-of-the-art performance on ImageNet VID dataset. Code is available at \url{https://github.com/Scalsol/mega.pytorch}.
Tasks Object Detection, Video Object Detection
Published 2020-03-26
URL https://arxiv.org/abs/2003.12063v1
PDF https://arxiv.org/pdf/2003.12063v1.pdf
PWC https://paperswithcode.com/paper/memory-enhanced-global-local-aggregation-for
Repo https://github.com/Scalsol/mega.pytorch
Framework pytorch

Design, Validation, and Case Studies of 2D-VSR-Sim, an Optimization-friendly Simulator of 2-D Voxel-based Soft Robots

Title Design, Validation, and Case Studies of 2D-VSR-Sim, an Optimization-friendly Simulator of 2-D Voxel-based Soft Robots
Authors Eric Medvet, Alberto Bartoli, Andrea De Lorenzo, Stefano Seriani
Abstract Voxel-based soft robots (VSRs) are aggregations of soft blocks whose design is amenable to optimization. We here present a software, 2D-VSR-Sim, for facilitating research concerning the optimization of VSRs body and brain. The software, written in Java, provides consistent interfaces for all the VSRs aspects suitable for optimization and considers by design the presence of sensing, i.e., the possibility of exploiting the feedback from the environment for controlling the VSR. We experimentally characterize, from a mechanical point of view, the VSRs that can be simulated with 2D-VSR-Sim and we discuss the computational burden of the simulation. Finally, we show how 2D-VSR-Sim can be used to repeat the experiments of significant previous studies and, in perspective, to provide experimental answers to a variety of research questions.
Published 2020-01-23
URL https://arxiv.org/abs/2001.08617v2
PDF https://arxiv.org/pdf/2001.08617v2.pdf
PWC https://paperswithcode.com/paper/2d-vsr-sim-an-optimization-friendly-simulator
Repo https://github.com/ericmedvet/2dhmsr
Framework none

Efficient Rule Learning with Template Saturation for Knowledge Graph Completion

Title Efficient Rule Learning with Template Saturation for Knowledge Graph Completion
Authors Yulong Gu, Yu Guan, Paolo Missier
Abstract The logic-based methods that learn first-order rules from knowledge graphs (KGs) for knowledge graph completion (KGC) task are desirable in that the learnt models are inductive, interpretable and transferable. The challenge in such rule learners is that the expressive rules are often buried in vast rule space, and the procedure of identifying expressive rules by measuring rule quality is costly to execute. Therefore, optimizations on rule generation and evaluation are in need. In this work, we propose a novel bottom-up probabilistic rule learner that features: 1.) a two-stage procedure for optimized rule generation where the system first generalizes paths sampled from a KG into template rules that contain no constants until a certain degree of template saturation is achieved and then specializes template rules into instantiated rules that contain constants; 2.) a grouping technique for optimized rule evaluation where structurally similar instantiated rules derived from the same template rules are put into the same groups and evaluated collectively over the groundings of the deriving template rules. Through extensive experiments over large benchmark datasets on KGC task, our algorithm demonstrates consistent and substantial performance improvements over all of the state-of-the-art baselines.
Tasks Knowledge Graph Completion, Knowledge Graphs, Link Prediction
Published 2020-03-13
URL https://arxiv.org/abs/2003.06071v1
PDF https://arxiv.org/pdf/2003.06071v1.pdf
PWC https://paperswithcode.com/paper/efficient-rule-learning-with-template
Repo https://github.com/irokin/GPFL
Framework none

Temporal Attribute Prediction via Joint Modeling of Multi-Relational Structure Evolution

Title Temporal Attribute Prediction via Joint Modeling of Multi-Relational Structure Evolution
Authors Sankalp Garg, Navodita Sharma, Woojeong Jin, Xiang Ren
Abstract Time series prediction is an important problem in machine learning. Previous methods for time series prediction did not involve additional information. With a lot of dynamic knowledge graphs available, we can use this additional information to predict the time series better. Recently, there has been a focus on the application of deep representation learning on dynamic graphs. These methods predict the structure of the graph by reasoning over the interactions in the graph at previous time steps. In this paper, we propose a new framework to incorporate the information from dynamic knowledge graphs for time series prediction. We show that if the information contained in the graph and the time series data are closely related, then this inter-dependence can be used to predict the time series with improved accuracy. Our framework, DArtNet, learns a static embedding for every node in the graph as well as a dynamic embedding which is dependent on the dynamic attribute value (time-series). Then it captures the information from the neighborhood by taking a relation specific mean and encodes the history information using RNN. We jointly train the model link prediction and attribute prediction. We evaluate our method on five specially curated datasets for this problem and show a consistent improvement in time series prediction results.
Tasks Knowledge Graphs, Link Prediction, Representation Learning, Time Series, Time Series Prediction
Published 2020-03-09
URL https://arxiv.org/abs/2003.03919v1
PDF https://arxiv.org/pdf/2003.03919v1.pdf
PWC https://paperswithcode.com/paper/temporal-attribute-prediction-via-joint
Repo https://github.com/INK-USC/DArtNet
Framework pytorch

Epileptic Seizure Classification with Symmetric and Hybrid Bilinear Models

Title Epileptic Seizure Classification with Symmetric and Hybrid Bilinear Models
Authors Tennison Liu, Nhan Duy Truong, Armin Nikpour, Luping Zhou, Omid Kavehei
Abstract Epilepsy affects nearly 1% of the global population, of which two thirds can be treated by anti-epileptic drugs and a much lower percentage by surgery. Diagnostic procedures for epilepsy and monitoring are highly specialized and labour-intensive. The accuracy of the diagnosis is also complicated by overlapping medical symptoms, varying levels of experience and inter-observer variability among clinical professions. This paper proposes a novel hybrid bilinear deep learning network with an application in the clinical procedures of epilepsy classification diagnosis, where the use of surface electroencephalogram (sEEG) and audiovisual monitoring is standard practice. Hybrid bilinear models based on two types of feature extractors, namely Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), are trained using Short-Time Fourier Transform (STFT) of one-second sEEG. In the proposed hybrid models, CNNs extract spatio-temporal patterns, while RNNs focus on the characteristics of temporal dynamics in relatively longer intervals given the same input data. Second-order features, based on interactions between these spatio-temporal features are further explored by bilinear pooling and used for epilepsy classification. Our proposed methods obtain an F1-score of 97.4% on the Temple University Hospital Seizure Corpus and 97.2% on the EPILEPSIAE dataset, comparing favourably to existing benchmarks for sEEG-based seizure type classification. The open-source implementation of this study is available at https://github.com/NeuroSyd/Epileptic-Seizure-Classification
Published 2020-01-15
URL https://arxiv.org/abs/2001.06282v1
PDF https://arxiv.org/pdf/2001.06282v1.pdf
PWC https://paperswithcode.com/paper/epileptic-seizure-classification-with
Repo https://github.com/NeuroSyd/Epileptic-Seizure-Classification
Framework none

Think Global, Act Local: Relating DNN generalisation and node-level SNR

Title Think Global, Act Local: Relating DNN generalisation and node-level SNR
Authors Paul Norridge
Abstract The reasons behind good DNN generalisation remain an open question. In this paper we explore the problem by looking at the Signal-to-Noise Ratio of nodes in the network. Starting from information theory principles, it is possible to derive an expression for the SNR of a DNN node output. Using this expression we construct figures-of-merit that quantify how well the weights of a node optimise SNR (or, equivalently, information rate). Applying these figures-of-merit, we give examples indicating that weight sets that promote good SNR performance also exhibit good generalisation. In addition, we are able to identify the qualities of weight sets that exhibit good SNR behaviour and hence promote good generalisation. This leads to a discussion of how these results relate to network training and regularisation. Finally, we identify some ways that these observations can be used in training design.
Published 2020-02-11
URL https://arxiv.org/abs/2002.04687v1
PDF https://arxiv.org/pdf/2002.04687v1.pdf
PWC https://paperswithcode.com/paper/think-global-act-local-relating-dnn
Repo https://github.com/pnorridge/think-global-act-local
Framework none

PAPRIKA: Private Online False Discovery Rate Control

Title PAPRIKA: Private Online False Discovery Rate Control
Authors Wanrong Zhang, Gautam Kamath, Rachel Cummings
Abstract In hypothesis testing, a false discovery occurs when a hypothesis is incorrectly rejected due to noise in the sample. When adaptively testing multiple hypotheses, the probability of a false discovery increases as more tests are performed. Thus the problem of False Discovery Rate (FDR) control is to find a procedure for testing multiple hypotheses that accounts for this effect in determining the set of hypotheses to reject. The goal is to minimize the number (or fraction) of false discoveries, while maintaining a high true positive rate (i.e., correct discoveries). In this work, we study False Discovery Rate (FDR) control in multiple hypothesis testing under the constraint of differential privacy for the sample. Unlike previous work in this direction, we focus on the online setting, meaning that a decision about each hypothesis must be made immediately after the test is performed, rather than waiting for the output of all tests as in the offline setting. We provide new private algorithms based on state-of-the-art results in non-private online FDR control. Our algorithms have strong provable guarantees for privacy and statistical performance as measured by FDR and power. We also provide experimental results to demonstrate the efficacy of our algorithms in a variety of data environments.
Published 2020-02-27
URL https://arxiv.org/abs/2002.12321v1
PDF https://arxiv.org/pdf/2002.12321v1.pdf
PWC https://paperswithcode.com/paper/paprika-private-online-false-discovery-rate
Repo https://github.com/wanrongz/PAPRIKA
Framework none

Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)

Title Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
Authors Yasir Alanazi, N. Sato, Tianbo Liu, W. Melnitchouk, Michelle P. Kuchera, Evan Pritchard, Michael Robertson, Ryan Strauss, Luisa Velasco, Yaohang Li
Abstract We apply generative adversarial network (GAN) technology to build an event generator that simulates particle production in electron-proton scattering that is free of theoretical assumptions about underlying particle dynamics. The difficulty of efficiently training a GAN event simulator lies in learning the complicated patterns of the distributions of the particles physical properties. We develop a GAN that selects a set of transformed features from particle momenta that can be generated easily by the generator, and uses these to produce a set of augmented features that improve the sensitivity of the discriminator. The new Feature-Augmented and Transformed GAN (FAT-GAN) is able to faithfully reproduce the distribution of final state electron momenta in inclusive electron scattering, without the need for input derived from domain-based theoretical assumptions. The developed technology can play a significant role in boosting the science of the Jefferson Lab 12 GeV program and the future Electron-Ion Collider.
Published 2020-01-29
URL https://arxiv.org/abs/2001.11103v1
PDF https://arxiv.org/pdf/2001.11103v1.pdf
PWC https://paperswithcode.com/paper/simulation-of-electron-proton-scattering
Repo https://github.com/JeffersonLab/FAT-GAN
Framework none

Invertible Generative Modeling using Linear Rational Splines

Title Invertible Generative Modeling using Linear Rational Splines
Authors Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie
Abstract Normalizing flows attempt to model an arbitrary probability distribution through a set of invertible mappings. These transformations are required to achieve a tractable Jacobian determinant that can be used in high-dimensional scenarios. The first normalizing flow designs used coupling layer mappings built upon affine transformations. The significant advantage of such models is their easy-to-compute inverse. Nevertheless, making use of affine transformations may limit the expressiveness of such models. Recently, invertible piecewise polynomial functions as a replacement for affine transformations have attracted attention. However, these methods require solving a polynomial equation to calculate their inverse. In this paper, we explore using linear rational splines as a replacement for affine transformations used in coupling layers. Besides having a straightforward inverse, inference and generation have similar cost and architecture in this method. Moreover, simulation results demonstrate the competitiveness of this approach’s performance compared to existing methods.
Published 2020-01-15
URL https://arxiv.org/abs/2001.05168v3
PDF https://arxiv.org/pdf/2001.05168v3.pdf
PWC https://paperswithcode.com/paper/invertible-generative-modeling-using-linear
Repo https://github.com/hmdolatabadi/LRS_NF
Framework pytorch

(De)Randomized Smoothing for Certifiable Defense against Patch Attacks

Title (De)Randomized Smoothing for Certifiable Defense against Patch Attacks
Authors Alexander Levine, Soheil Feizi
Abstract Patch adversarial attacks on images, in which the attacker can distort pixels within a region of bounded size, are an important threat model since they provide a quantitative model for physical adversarial attacks. In this paper, we introduce a certifiable defense against patch attacks that guarantees for a given image and patch attack size, no patch adversarial examples exist. Our method is related to the broad class of randomized smoothing robustness schemes which provide high-confidence probabilistic robustness certificates. By exploiting the fact that patch attacks are more constrained than general sparse attacks, we derive meaningfully large robustness certificates. Additionally, the algorithm we propose is de-randomized, providing deterministic certificates. To the best of our knowledge, there exists only one prior method for certifiable defense against patch attacks, which relies on interval bound propagation. While this sole existing method performs well on MNIST, it has several limitations: it requires computationally expensive training, does not scale to ImageNet, and performs poorly on CIFAR-10. In contrast, our proposed method effectively addresses all of these issues: our classifier can be trained quickly, achieves high clean and certified robust accuracy on CIFAR-10, and provides certificates at the ImageNet scale. For example, for a 5*5 patch attack on CIFAR-10, our method achieves up to around 57.8% certified accuracy (with a classifier around 83.9% clean accuracy), compared to at most 30.3% certified accuracy for the existing method (with a classifier with around 47.8% clean accuracy), effectively establishing a new state-of-the-art. Code is available at https://github.com/alevine0/patchSmoothing.
Published 2020-02-25
URL https://arxiv.org/abs/2002.10733v1
PDF https://arxiv.org/pdf/2002.10733v1.pdf
PWC https://paperswithcode.com/paper/derandomized-smoothing-for-certifiable
Repo https://github.com/alevine0/patchSmoothing
Framework pytorch

Personalized Activity Recognition with Deep Triplet Embeddings

Title Personalized Activity Recognition with Deep Triplet Embeddings
Authors David M. Burns, Cari M. Whyne
Abstract A significant challenge for a supervised learning approach to inertial human activity recognition is the heterogeneity of data between individual users, resulting in very poor performance of impersonal algorithms for some subjects. We present an approach to personalized activity recognition based on deep embeddings derived from a fully convolutional neural network. We experiment with both categorical cross entropy loss and triplet loss for training the embedding, and describe a novel triplet loss function based on subject triplets. We evaluate these methods on three publicly available inertial human activity recognition data sets (MHEALTH, WISDM, and SPAR) comparing classification accuracy, out-of-distribution activity detection, and embedding generalization to new activities. The novel subject triplet loss provides the best performance overall, and all personalized deep embeddings out-perform our baseline personalized engineered feature embedding and an impersonal fully convolutional neural network classifier.
Tasks Action Detection, Activity Detection, Activity Recognition, Human Activity Recognition
Published 2020-01-15
URL https://arxiv.org/abs/2001.05517v1
PDF https://arxiv.org/pdf/2001.05517v1.pdf
PWC https://paperswithcode.com/paper/personalized-activity-recognition-with-deep
Repo https://github.com/dmbee/fcn-core
Framework none

Generalizable Semantic Segmentation via Model-agnostic Learning and Target-specific Normalization

Title Generalizable Semantic Segmentation via Model-agnostic Learning and Target-specific Normalization
Authors Jian Zhang, Lei Qi, Yinghuan Shi, Yang Gao
Abstract Semantic segmentation methods in the supervised scenario have achieved significant improvement in recent years. However, when directly deploying the trained model to segment the images of unseen (or new coming) domains, its performance usually drops dramatically due to the data-distribution discrepancy between seen and unseen domains. To overcome this limitation, we propose a novel domain generalization framework for the generalizable semantic segmentation task, which enhances the generalization ability of the model from two different views, including the training paradigm and the data-distribution discrepancy. Concretely, we exploit the model-agnostic learning method to simulate the domain shift problem, which deals with the domain generalization from the training scheme perspective. Besides, considering the data-distribution discrepancy between source domains and unseen target domains, we develop the target-specific normalization scheme to further boost the generalization ability in unseen target domains. Extensive experiments highlight that the proposed method produces state-of-the-art performance for the domain generalization of semantic segmentation on multiple benchmark segmentation datasets (i.e., Cityscapes, Mapillary). Furthermore, we gain an interesting observation that the target-specific normalization can benefit from the model-agnostic learning scheme.
Tasks Domain Generalization, Semantic Segmentation
Published 2020-03-27
URL https://arxiv.org/abs/2003.12296v1
PDF https://arxiv.org/pdf/2003.12296v1.pdf
PWC https://paperswithcode.com/paper/generalizable-semantic-segmentation-via-model
Repo https://github.com/koncle/TSMLDG
Framework pytorch

Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection

Title Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection
Authors Jianyuan Guo, Kai Han, Yunhe Wang, Chao Zhang, Zhaohui Yang, Han Wu, Xinghao Chen, Chang Xu
Abstract Neural Architecture Search (NAS) has achieved great success in image classification task. Some recent works have managed to explore the automatic design of efficient backbone or feature fusion layer for object detection. However, these methods focus on searching only one certain component of object detector while leaving others manually designed. We identify the inconsistency between searched component and manually designed ones would withhold the detector of stronger performance. To this end, we propose a hierarchical trinity search framework to simultaneously discover efficient architectures for all components (i.e. backbone, neck, and head) of object detector in an end-to-end manner. In addition, we empirically reveal that different parts of the detector prefer different operators. Motivated by this, we employ a novel scheme to automatically screen different sub search spaces for different components so as to perform the end-to-end search for each component on the corresponding sub search space efficiently. Without bells and whistles, our searched architecture, namely Hit-Detector, achieves 41.4% mAP on COCO minival set with 27M parameters. Our implementation is available at https://github.com/ggjy/HitDet.pytorch.
Tasks Image Classification, Neural Architecture Search, Object Detection
Published 2020-03-26
URL https://arxiv.org/abs/2003.11818v1
PDF https://arxiv.org/pdf/2003.11818v1.pdf
PWC https://paperswithcode.com/paper/hit-detector-hierarchical-trinity
Repo https://github.com/ggjy/HitDet.pytorch
Framework pytorch

Lifelong Learning with Searchable Extension Units

Title Lifelong Learning with Searchable Extension Units
Authors Wenjin Wang, Yunqing Hu, Yin Zhang
Abstract Lifelong learning remains an open problem. One of its main difficulties is catastrophic forgetting. Many dynamic expansion approaches have been proposed to address this problem, but they all use homogeneous models of predefined structure for all tasks. The common original model and expansion structures ignore the requirement of different model structures on different tasks, which leads to a less compact model for multiple tasks and causes the model size to increase rapidly as the number of tasks increases. Moreover, they can not perform best on all tasks. To solve those problems, in this paper, we propose a new lifelong learning framework named Searchable Extension Units (SEU) by introducing Neural Architecture Search into lifelong learning, which breaks down the need for a predefined original model and searches for specific extension units for different tasks, without compromising the performance of the model on different tasks. Our approach can obtain a much more compact model without catastrophic forgetting. The experimental results on the PMNIST, the split CIFAR10 dataset, the split CIFAR100 dataset, and the Mixture dataset empirically prove that our method can achieve higher accuracy with much smaller model, whose size is about 25-33 percentage of that of the state-of-the-art methods.
Tasks Neural Architecture Search
Published 2020-03-19
URL https://arxiv.org/abs/2003.08559v1
PDF https://arxiv.org/pdf/2003.08559v1.pdf
PWC https://paperswithcode.com/paper/lifelong-learning-with-searchable-extension
Repo https://github.com/WenjinW/LLSEU
Framework pytorch
comments powered by Disqus