2020.07.15.txt

==========New Papers==========
1, TITLE:       Lower Bounds of Algebraic Branching Programs and Layerization
http://arxiv.org/abs/2007.06819
AUTHORS:        Christian Engels
HIGHLIGHT:      In this paper we improve the lower bound of Chatterjee et al.\ (ECCC 2019) to an $\Omega(n^2)$ lower bound for unlayered Algebraic Branching Programs.

2, TITLE:       Improving Pixel Embedding Learning through Intermediate Distance Regression Supervision for Instance Segmentation
http://arxiv.org/abs/2007.06660
AUTHORS:        Yuli Wu ; Long Chen ; Dorit Merhof
COMMENTS:       ECCV 2020 Workshop: Computer Vision Problems in Plant Phenotyping (CVPPP 2020)
HIGHLIGHT:      In this work, we propose a simple, yet highly effective, architecture for object-aware embedding learning.

3, TITLE:       What's in a Name? Are BERT Named Entity Representations just as Good for any other Name?
http://arxiv.org/abs/2007.06897
AUTHORS:        Sriram Balasubramanian ; Naman Jain ; Gaurav Jindal ; Abhijeet Awasthi ; Sunita Sarawagi
COMMENTS:       Accepted at RepL4NLP, ACL2020
HIGHLIGHT:      We evaluate named entity representations of BERT-based NLP models by investigating their robustness to replacements from the same typed class in the input.

4, TITLE:       Closed-Form Factorization of Latent Semantics in GANs
http://arxiv.org/abs/2007.06600
AUTHORS:        Yujun Shen ; Bolei Zhou
COMMENTS:       11 pages, 8 figures, 3 tables
HIGHLIGHT:      In this work, we examine the internal representation learned by GANs to reveal the underlying variation factors in an unsupervised manner.

5, TITLE:       Face to Purchase: Predicting Consumer Choices with Structured Facial and Behavioral Traits Embedding
http://arxiv.org/abs/2007.06842
AUTHORS:        Zhe Liu ; Xianzhi Wang ; Lina Yao ; Jake An ; Lei Bai ; Ee-Peng Lim
HIGHLIGHT:      We propose to predict consumers' purchases based on their facial features and purchasing histories.

6, TITLE:       Our Evaluation Metric Needs an Update to Encourage Generalization
http://arxiv.org/abs/2007.06898
AUTHORS:        Swaroop Mishra ; Anjana Arunkumar ; Chris Bryan ; Chitta Baral
COMMENTS:       Accepted to ICML UDL 2020
HIGHLIGHT:      In order to stop the inflation in model performance -- and thus overestimation in AI systems' capabilities -- we propose a simple and novel evaluation metric, WOOD Score, that encourages generalization during evaluation.

7, TITLE:       UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a Generic Framework for Handling Common Camera Distortion Models
http://arxiv.org/abs/2007.06676
AUTHORS:        Varun Ravi Kumar ; Senthil Yogamani ; Markus Bach ; Christian Witt ; Stefan Milz ; Patrick Mader
COMMENTS:       Accepted for Presentation at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020
HIGHLIGHT:      In this paper, we propose a generic scale-aware self-supervised pipeline for estimating depth, euclidean distance, and visual odometry from unrectified monocular videos.

8, TITLE:       Socially and Contextually Aware Human Motion and Pose Forecasting
http://arxiv.org/abs/2007.06843
AUTHORS:        Vida Adeli ; Ehsan Adeli ; Ian Reid ; Juan Carlos Niebles ; Hamid Rezatofighi
COMMENTS:       Accepted in RA-L and IROS
HIGHLIGHT:      In this paper, we propose a novel framework to tackle both tasks of human motion (or trajectory) and body skeleton pose forecasting in a unified end-to-end pipeline.

9, TITLE:       Landslide Segmentation with U-Net: Evaluating Different Sampling Methods and Patch Sizes
http://arxiv.org/abs/2007.06672
AUTHORS:        Lucas P. Soares ; Helen C. Dias ; Carlos H. Grohmann
COMMENTS:       13 pages, 7 figures, 3 tables
HIGHLIGHT:      The results achieved in this study are comparable to other fully convolutional models found in the literature, increasing the knowledge in the area.

10, TITLE:      Learning Differential Diagnosis of Skin Conditions with Co-occurrence Supervision using Graph Convolutional Networks
http://arxiv.org/abs/2007.06666
AUTHORS:        Junyan Wu ; Hao Jiang ; Xiaowei Ding ; Anudeep Konda ; Jin Han ; Yang Zhang ; Qian Li
HIGHLIGHT:      In this paper, we propose a deep learning system (DLS) that may predict differential diagnosis of skin conditions using clinical images.

11, TITLE:      Deep Heterogeneous Autoencoder for Subspace Clustering of Sequential Data
http://arxiv.org/abs/2007.07175
AUTHORS:        Abubakar Siddique ; Reza Jalil Mozhdehi ; Henry Medeiros
COMMENTS:       14 pages, 3 figures, submitted to accv
HIGHLIGHT:      We propose an unsupervised learning approach using a convolutional and fully connected autoencoder, which we call deep heterogeneous autoencoder, to learn discriminative features from segmentation masks and detection bounding boxes.

12, TITLE:      Conditional Image Retrieval
http://arxiv.org/abs/2007.07177
AUTHORS:        Mark Hamilton ; Stephanie Fu ; William T. Freeman ; Mindren Lu
HIGHLIGHT:      As a motivating application, we present an algorithm that can explore shared semantic content between works of art of vastly different media and cultural origin. We present two new datasets for evaluating the performance of CIR systems and evaluate a variety of design choices.

13, TITLE:      Towards Dense People Detection with Deep Learning and Depth images
http://arxiv.org/abs/2007.07171
AUTHORS:        David Fuentes-Jimenez ; Cristina Losada-Gutierrez ; David Casillas-Perez ; Javier Macias-Guarasa ; Roberto Martin-Lopez ; Daniel Pizarro ; Carlos A. Luna
HIGHLIGHT:      This paper proposes a DNN-based system that detects multiple people from a single depth image.

14, TITLE:      Wavelet-Based Dual-Branch Network for Image Demoireing
http://arxiv.org/abs/2007.07173
AUTHORS:        Lin Liu ; Jianzhuang Liu ; Shanxin Yuan ; Gregory Slabaugh ; Ales Leonardis ; Wengang Zhou ; Qi Tian
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this paper, we design a wavelet-based dual-branch network (WDNet) with a spatial attention mechanism for image demoireing.

15, TITLE:      Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization
http://arxiv.org/abs/2007.06643
AUTHORS:        Kyle Min ; Jason J. Corso
COMMENTS:       ECCV 2020 camera ready (Supplementary material: on ECVA soon)
HIGHLIGHT:      To address this issue, we propose a novel method named A2CL-PT.

16, TITLE:      Goal-Aware Prediction: Learning to Model What Matters
http://arxiv.org/abs/2007.07170
AUTHORS:        Suraj Nair ; Silvio Savarese ; Chelsea Finn
HIGHLIGHT:      In this paper, we propose to direct prediction towards task relevant information, enabling the model to be aware of the current task and encouraging it to only model relevant quantities of the state space, resulting in a learning objective that more closely matches the downstream task.

17, TITLE:      Conditional Independences and Causal Relations implied by Sets of Equations
http://arxiv.org/abs/2007.07183
AUTHORS:        Tineke Blom ; Mirthe M. van Diepen ; Joris M. Mooij
HIGHLIGHT:      We prove that, under a solvability assumption, we can construct a Markov ordering graph that implies conditional independences and a causal ordering graph that encodes the effects of soft and perfect interventions by making use of Simon's causal ordering algorithm.

18, TITLE:      Compare and Reweight: Distinctive Image Captioning Using Similar Images Sets
http://arxiv.org/abs/2007.06877
AUTHORS:        Jiuniu Wang ; Wenjia Xu ; Qingzhong Wang ; Antoni B. Chan
HIGHLIGHT:      In this paper, we aim to improve the distinctiveness of image captions through training with sets of similar images.

19, TITLE:      Component Order Connectivity in Directed Graphs
http://arxiv.org/abs/2007.06896
AUTHORS:        J. Bang-Jensen ; E. Eiben ; G. Gutin ; M. Wahlstrom ; A. Yeo
HIGHLIGHT:      $ We study parametered complexity of DCOC for general and semicomplete digraphs with the following parameters: $k, \ell,\ell+k$ and $n-\ell$.

20, TITLE:      Joint Layout Analysis, Character Detection and Recognition for Historical Document Digitization
http://arxiv.org/abs/2007.06890
AUTHORS:        Weihong Ma ; Hesuo Zhang ; Lianwen Jin ; Sihang Wu ; Jiapeng Wang ; Yongpan Wang
COMMENTS:       6 pages, 6 figures
HIGHLIGHT:      In this paper, we propose an end-to-end trainable framework for restoring historical documents content that follows the correct reading order.

21, TITLE:      360$^\circ$ Depth Estimation from Multiple Fisheye Images with Origami Crown Representation of Icosahedron
http://arxiv.org/abs/2007.06891
AUTHORS:        Ren Komatsu ; Hiromitsu Fujii ; Yusuke Tamura ; Atsushi Yamashita ; Hajime Asama
COMMENTS:       8 pages, Accepted to the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2020). For supplementary video, see https://youtu.be/_vVD-zDMvyM
HIGHLIGHT:      In this study, we present a method for all-around depth estimation from multiple omnidirectional images for indoor environments.

22, TITLE:      The Collatz process embeds a base conversion algorithm
http://arxiv.org/abs/2007.06979
AUTHORS:        Tristan Stérin ; Damien Woods
COMMENTS:       8 figures
HIGHLIGHT:      We define a quasi-cellular automaton that exactly simulates the Collatz process on the square grid: on input $x\in\mathbb{N}$, written horizontally in base 2, successive rows give the Collatz sequence of $x$ in base 2.

23, TITLE:      Extracting Structured Data from Physician-Patient Conversations By Predicting Noteworthy Utterances
http://arxiv.org/abs/2007.07151
AUTHORS:        Kundan Krishna ; Amy Pavel ; Benjamin Schloss ; Jeffrey P. Bigham ; Zachary C. Lipton
HIGHLIGHT:      In this paper, we leverage this data to extract structured information that might assist physicians with post-visit documentation in electronic health records, potentially lightening the clerical burden.

24, TITLE:      Improving Face Recognition by Clustering Unlabeled Faces in the Wild
http://arxiv.org/abs/2007.06995
AUTHORS:        Aruni RoyChowdhury ; Xiang Yu ; Kihyuk Sohn ; Erik Learned-Miller ; Manmohan Chandraker
COMMENTS:       ECCV 2020
HIGHLIGHT:      To address this, we propose a novel identity separation method based on extreme value theory.

25, TITLE:      From Symmetry to Geometry: Tractable Nonconvex Problems
http://arxiv.org/abs/2007.06753
AUTHORS:        Yuqian Zhang ; Qing Qu ; John Wright
COMMENTS:       review paper, 33 pages, 10 figures
HIGHLIGHT:      The goal of this survey is to highlight a class of tractable nonconvex problems, which can be understood through the lens of symmetries.

26, TITLE:      Water level prediction from social media images with a multi-task ranking approach
http://arxiv.org/abs/2007.06749
AUTHORS:        P. Chaudhary ; S. D'Aronco ; J. P. Leitao ; K. Schindler ; J. D. Wegner
COMMENTS:       Accepted in ISPRS Journal 2020
HIGHLIGHT:      We propose a multi-task (deep) learning approach, where a model is trained using both a regression and a pairwise ranking loss. Moreover, we provide a new dataset, named DeepFlood, with 8145 annotated ground-level images, and show that the proposed multi-task approach can predict the water level from a single, crowd-sourced image with ~11 cm root mean square error.

27, TITLE:      An Uncertainty-based Human-in-the-loop System for Industrial Tool Wear Analysis
http://arxiv.org/abs/2007.07129
AUTHORS:        Alexander Treiss ; Jannis Walk ; Niklas Kühl
COMMENTS:       Alexander Treiss and Jannis Walk contributed equally in shared first authorship
HIGHLIGHT:      To address these issues, we use uncertainty measures based on Monte-Carlo dropout in the context of a human-in-the-loop system to increase the system's transparency and performance.

28, TITLE:      Correlation filter tracking with adaptive proposal selection for accurate scale estimation
http://arxiv.org/abs/2007.07018
AUTHORS:        Luo Xiong ; Yanjie Liang ; Yan Yan ; Hanzi Wang
COMMENTS:       6 pages, 14 figures
HIGHLIGHT:      In this paper, we propose an adaptive proposal selection algorithm which can generate a small number of high-quality proposals to handle the problem of scale variations for visual object tracking.

29, TITLE:      Multitask Learning Strengthens Adversarial Robustness
http://arxiv.org/abs/2007.07236
AUTHORS:        Chengzhi Mao ; Amogh Gupta ; Vikram Nitin ; Baishakhi Ray ; Shuran Song ; Junfeng Yang ; Carl Vondrick
HIGHLIGHT:      We present both theoretical and empirical analyses that connect the adversarial robustness of a model to the number of tasks that it is trained on.

30, TITLE:      Cross-Domain Medical Image Translation by Shared Latent Gaussian Mixture Model
http://arxiv.org/abs/2007.07230
AUTHORS:        Yingying Zhu ; Youbao Tang ; Yuxing Tang ; Daniel C. Elton ; Sungwon Lee ; Perry J. Pickhardt ; Ronald M. Summers
COMMENTS:       Accepted to Medical Image Computing and Computer Assisted Intervention (MICCAI) 2020
HIGHLIGHT:      In order to preserve fine structures during medical image translation, we propose a patch-based model using shared latent variables from a Gaussian mixture model.

31, TITLE:      Deployment and Evaluation of a Flexible Human-Robot Collaboration Model Based on AND/OR Graphs in a Manufacturing Environment
http://arxiv.org/abs/2007.06720
AUTHORS:        Prajval Kumar Murali ; Kourosh Darvish ; Fulvio Mastrogiovanni
HIGHLIGHT:      The results of this user study demonstrate how human-robot collaboration models like the one we propose can leverage the flexibility and the comfort of operators in the workplace.

32, TITLE:      Learning Semantics-enriched Representation via Self-discovery, Self-classification, and Self-restoration
http://arxiv.org/abs/2007.06959
AUTHORS:        Fatemeh Haghighi ; Mohammad Reza Hosseinzadeh Taher ; Zongwei Zhou ; Michael B. Gotway ; Jianming Liang
COMMENTS:       International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2020)
HIGHLIGHT:      We examine our Semantic Genesis with all the publicly-available pre-trained models, by either self-supervision or fully supervision, on the six distinct target tasks, covering both classification and segmentation in various medical modalities (i.e.,CT, MRI, and X-ray).

33, TITLE:      DETCID: Detection of Elongated Touching Cells with Inhomogeneous Illumination using a Deep Adversarial Network
http://arxiv.org/abs/2007.06716
AUTHORS:        Ali Memariani ; Ioannis A. Kakadiaris
HIGHLIGHT:      In this paper, DETCID, a deep cell detection method using adversarial training, specifically robust to inhomogeneous illumination and occlusion, is proposed.

34, TITLE:      COVID-19 Twitter Dataset with Latent Topics, Sentiments and Emotions Attributes
http://arxiv.org/abs/2007.06954
AUTHORS:        Raj Kumar Gupta ; Ajay Vishwanath ; Yinping Yang
COMMENTS:       20 pages, 5 figures, 9 tables
HIGHLIGHT:      This resource paper describes a large dataset covering over 63 million coronavirus-related Twitter posts from more than 13 million unique users since 28 January to 1 July 2020.

35, TITLE:      Multiview Detection with Feature Perspective Transformation
http://arxiv.org/abs/2007.07247
AUTHORS:        Yunzhong Hou ; Liang Zheng ; Stephen Gould
HIGHLIGHT:      To address these questions, we propose a novel multiview detection system, MVDet.

36, TITLE:      Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter
http://arxiv.org/abs/2007.07243
AUTHORS:        Guilin Liu ; Rohan Taori ; Ting-Chun Wang ; Zhiding Yu ; Shiqiu Liu ; Fitsum A. Reda ; Karan Sapra ; Andrew Tao ; Bryan Catanzaro
HIGHLIGHT:      In this work, based on the discovery that the assembling/stitching operation in traditional texture synthesis is analogous to a transposed convolution operation, we propose a novel way of using transposed convolution operation.

37, TITLE:      Inertial Sensing Meets Artificial Intelligence: Opportunity or Challenge?
http://arxiv.org/abs/2007.06727
AUTHORS:        You Li ; Ruizhi Chen ; Xiaoji Niu ; Yuan Zhuang ; Zhouzheng Gao ; Xin Hu ; Naser El-Sheimy
HIGHLIGHT:      This article reviews the research on using AI technology to enhance inertial sensing from various aspects, including sensor design and selection, calibration and error modeling, navigation and motion-sensing algorithms, multi-sensor information fusion, system evaluation, and practical application.

38, TITLE:      Modeling Artistic Workflows for Image Generation and Editing
http://arxiv.org/abs/2007.07238
AUTHORS:        Hung-Yu Tseng ; Matthew Fisher ; Jingwan Lu ; Yijun Li ; Vladimir Kim ; Ming-Hsuan Yang
COMMENTS:       ECCV 2020. Code: https://github.com/hytseng0509/ArtEditing
HIGHLIGHT:      Motivated by the above observations, we propose a generative model that follows a given artistic workflow, enabling both multi-stage image generation as well as multi-stage image editing of an existing piece of art.

39, TITLE:      P-KDGAN: Progressive Knowledge Distillation with GANs for One-class Novelty Detection
http://arxiv.org/abs/2007.06963
AUTHORS:        Zhiwei Zhang ; Shifeng Chen ; Lei Sun
COMMENTS:       IJCAI 2020
HIGHLIGHT:      In this paper, the Generative Adversarial Networks (GANs) based on encoder-decoder-encoder pipeline are used for detection and achieve state-of-the-art performance.

40, TITLE:      Automated Synthetic-to-Real Generalization
http://arxiv.org/abs/2007.06965
AUTHORS:        Wuyang Chen ; Zhiding Yu ; Zhangyang Wang ; Anima Anandkumar
COMMENTS:       Accepted to ICML 2020
HIGHLIGHT:      In this work, we explicitly encourage the synthetically trained model to maintain similar representations with the ImageNet pre-trained model, and propose a \textit{learning-to-optimize (L2O)} strategy to automate the selection of layer-wise learning rates.

41, TITLE:      Multi-Task Reinforcement Learning as a Hidden-Parameter Block MDP
http://arxiv.org/abs/2007.07206
AUTHORS:        Amy Zhang ; Shagun Sodhani ; Khimya Khetarpal ; Joelle Pineau
COMMENTS:       Accepted at the ICML 2020 Workshop on Theoretical Foundations of Reinforcement Learning
HIGHLIGHT:      In this work, we leverage ideas of common structure underlying a family of Markov decision processes (MDPs) to improve performance in the few-shot regime.

42, TITLE:      Visual Tracking by TridentAlign and Context Embedding
http://arxiv.org/abs/2007.06887
AUTHORS:        Janghoon Choi ; Junseok Kwon ; Kyoung Mu Lee
COMMENTS:       Code available on https://github.com/JanghoonChoi/TACT
HIGHLIGHT:      To address these persisting issues, we propose novel TridentAlign and context embedding modules for Siamese network-based visual tracking methods.

43, TITLE:      JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds
http://arxiv.org/abs/2007.06888
AUTHORS:        Zeyu Hu ; Mingmin Zhen ; Xuyang Bai ; Hongbo Fu ; Chiew-lan Tai
COMMENTS:       Accepted to ECCV 2020, supplementary materials included
HIGHLIGHT:      In this paper, we tackle the 3D semantic edge detection task for the first time and present a new two-stream fully-convolutional network that jointly performs the two tasks.

44, TITLE:      Knowledge Distillation for Multi-task Learning
http://arxiv.org/abs/2007.06889
AUTHORS:        Wei-Hong Li ; Hakan Bilen
COMMENTS:       We propose a knowledge distillation method for addressing the imbalance problem in multi-task learning
HIGHLIGHT:      To address the imbalance problem, we propose a knowledge distillation based method in this work.

45, TITLE:      AUTO3D: Novel view synthesis through unsupervisely learned variational viewpoint and global 3D representation
http://arxiv.org/abs/2007.06620
AUTHORS:        Xiaofeng Liu ; Tong Che ; Yiqun Lu ; Chao Yang ; Site Li ; Jane You
COMMENTS:       ECCV 2020
HIGHLIGHT:      In the viewer-centered coordinates, we construct an end-to-end trainable conditional variational framework to disentangle the unsupervisely learned relative-pose/rotation and implicit global 3D representation (shape, texture and the origin of viewer-centered coordinates, etc.).

46, TITLE:      Topology-Change-Aware Volumetric Fusion for Dynamic Scene Reconstruction
http://arxiv.org/abs/2007.06853
AUTHORS:        Chao Li ; Xiaohu Guo
COMMENTS:       European Conference on Computer Vision 2020
HIGHLIGHT:      In this paper, the classic framework is re-designed to enable 4D reconstruction of dynamic scene under topology changes, by introducing a novel structure of Non-manifold Volumetric Grid to the re-design of both TSDF and EDG, which allows connectivity updates by cell splitting and replication.

47, TITLE:      Inferring the 3D Standing Spine Posture from 2D Radiographs
http://arxiv.org/abs/2007.06612
AUTHORS:        Amirhossein Bayat ; Anjany Sekuboyina ; Johannes C. Paetzold ; Christian Payer ; Darko Stern ; Martin Urschler ; Jan S. Kirschke ; Bjoern H. Menze
HIGHLIGHT:      This work aims to integrate the two realms, i.e. it combines the upright spinal curvature from radiographs with the 3D vertebral shape from CT imaging for synthesizing an upright 3D model of spine, loaded naturally.

48, TITLE:      BUNET: Blind Medical Image Segmentation Based on Secure UNET
http://arxiv.org/abs/2007.06855
AUTHORS:        Song Bian ; Xiaowei Xu ; Weiwen Jiang ; Yiyu Shi ; Takashi Sato
COMMENTS:       11 pages, 2 figures, in Proceedings of International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2020)
HIGHLIGHT:      To ensure efficient machine learning as a service schemes while protecting data confidentiality, in this work, we propose blind UNET (BUNET), a secure protocol that implements privacy-preserving medical image segmentation based on the UNET architecture.

49, TITLE:      Dense Crowds Detection and Counting with a Lightweight Architecture
http://arxiv.org/abs/2007.06630
AUTHORS:        Javier Antonio Gonzalez-Trejo ; Diego Alberto Mercado-Ravell
HIGHLIGHT:      In this paper, we propose a lightweight convolutional neural network architecture to perform crowd detection and counting using fewer computer resources without a significant loss on count accuracy.

50, TITLE:      T-Basis: a Compact Representation for Neural Networks
http://arxiv.org/abs/2007.06631
AUTHORS:        Anton Obukhov ; Maxim Rakhuba ; Stamatios Georgoulis ; Menelaos Kanakis ; Dengxin Dai ; Luc Van Gool
COMMENTS:       Accepted at ICML 2020
HIGHLIGHT:      We introduce T-Basis, a novel concept for a compact representation of a set of tensors, each of an arbitrary shape, which is often seen in Neural Networks.

51, TITLE:      Robust Identifiability in Linear Structural Equation Models of Causal Inference
http://arxiv.org/abs/2007.06869
AUTHORS:        Karthik Abinav Sankararaman ; Anand Louis ; Navin Goyal
HIGHLIGHT:      In this work, we consider the problem of robust parameter estimation from observational data in the context of linear structural equation models (LSEMs).

52, TITLE:      Alleviating Over-segmentation Errors by Detecting Action Boundaries
http://arxiv.org/abs/2007.06866
AUTHORS:        Yuchi Ishikawa ; Seito Kasai ; Yoshimitsu Aoki ; Hirokatsu Kataoka
COMMENTS:       under review
HIGHLIGHT:      We propose an effective framework for the temporal action segmentation task, namely an Action Segment Refinement Framework (ASRF).

53, TITLE:      A new approach to descriptors generation for image retrieval by analyzing activations of deep neural network layers
http://arxiv.org/abs/2007.06624
AUTHORS:        Paweł Staszewski ; Maciej Jaworski ; Jinde Cao ; Leszek Rutkowski
COMMENTS:       8
HIGHLIGHT:      In this paper, we consider the problem of descriptors construction for the task of content-based image retrieval using deep neural networks.

54, TITLE:      Programming by Rewards
http://arxiv.org/abs/2007.06835
AUTHORS:        Nagarajan Natarajan ; Ajaykrishna Karthikeyan ; Prateek Jain ; Ivan Radicek ; Sriram Rajamani ; Sumit Gulwani ; Johannes Gehrke
HIGHLIGHT:      We present empirical evaluation against other baseline techniques over real-world case studies (including PROSE) as well on simple synthetic benchmarks.

55, TITLE:      Top-Related Meta-Learning Method for Few-Shot Detection
http://arxiv.org/abs/2007.06837
AUTHORS:        Qian Li ; Nan Guo ; Duo Wang ; Xiaochun Ye
COMMENTS:       Nips2020(desk rejected),meta-learing,few-shot detection
HIGHLIGHT:      Therefore, for meta-learning method of few-shot detection, we propose a TCL which exploits the true-label example and the most similar semantics with the example, and a category-based grouping mechanism which groups categories by appearance and environment to enhance the semantic features between similar categories.

56, TITLE:      Sudo rm -rf: Efficient Networks for Universal Audio Source Separation
http://arxiv.org/abs/2007.06833
AUTHORS:        Efthymios Tzinis ; Zhepei Wang ; Paris Smaragdis
COMMENTS:       accepted to MLSP 2020
HIGHLIGHT:      In this paper, we present an efficient neural network for end-to-end general purpose audio source separation.

57, TITLE:      A model to support collective reasoning: Formalization, analysis and computational assessment
http://arxiv.org/abs/2007.06850
AUTHORS:        Jordi Ganzer ; Natalia Criado ; Maite Lopez-Sanchez ; Simon Parsons ; Juan A. Rodriguez-Aguilar
HIGHLIGHT:      Inspired by e-participation systems, in this paper we propose a new model to represent human debates and methods to obtain collective conclusions from them.

58, TITLE:      Contextualized Code Representation Learning for Commit Message Generation
http://arxiv.org/abs/2007.06934
AUTHORS:        Lun Yiu Nie ; Cuiyun Gao ; Zhicong Zhong ; Wai Lam ; Yang Liu ; Zenglin Xu
HIGHLIGHT:      In this paper, we propose a novel Contextualized code representation learning method for commit message Generation (CoreGen).

59, TITLE:      Self-Supervised Monocular Depth Estimation: Solving the Dynamic Object Problem by Semantic Guidance
http://arxiv.org/abs/2007.06936
AUTHORS:        Marvin Klingner ; Jan-Aike Termöhlen ; Jonas Mikolajczyk ; Tim Fingscheidt
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work we present a new self-supervised semantically-guided depth estimation (SGDepth) method to deal with moving dynamic-class (DC) objects, such as moving cars and pedestrians, which violate the static-world assumptions typically made during training of such models.

60, TITLE:      REPrune: Filter Pruning via Representative Election
http://arxiv.org/abs/2007.06932
AUTHORS:        Mincheol Park ; Woojeong Kim ; Suhyun Kim
COMMENTS:       Under Review at ECCV 2020
HIGHLIGHT:      Our novel pruning method entitled "REPrune" addresses this problem by selecting representative filters via clustering.

61, TITLE:      Re-ranking for Writer Identification and Writer Retrieval
http://arxiv.org/abs/2007.07101
AUTHORS:        Simon Jordan ; Mathias Seuret ; Pavel Král ; Ladislav Lenc ; Jiří Martínek ; Barbara Wiermann ; Tobias Schwinger ; Andreas Maier ; Vincent Christlein
HIGHLIGHT:      We use these reciprocal relationships in two ways: encode them into new vectors, as originally proposed, or integrate them in terms of query-expansion.

62, TITLE:      Deep Transformer based Data Augmentation with Subword Units for Morphologically Rich Online ASR
http://arxiv.org/abs/2007.06949
AUTHORS:        Balázs Tarján ; György Szaszák ; Tibor Fegyó ; Péter Mihajlik
COMMENTS:       Submitted to Interspeech 2020
HIGHLIGHT:      Therefore, we propose a new method called subword-based neural text augmentation, where we retokenize the generated text into statistically derived subwords.

63, TITLE:      Modeling Voting for System Combination in Machine Translation
http://arxiv.org/abs/2007.06943
AUTHORS:        Xuancheng Huang ; Jiacheng Zhang ; Zhixing Tan ; Derek F. Wong ; Huanbo Luan ; Jingfang Xu ; Maosong Sun ; Yang Liu
COMMENTS:       Accepted by main track of IJCAI2020;SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence), all rights reserved. https://www.ijcai.org/Proceedings/2020/511
HIGHLIGHT:      In this work, we propose an approach to modeling voting for system combination in machine translation.

64, TITLE:      Fair Algorithms for Multi-Agent Multi-Armed Bandits
http://arxiv.org/abs/2007.06699
AUTHORS:        Safwan Hossain ; Evi Micha ; Nisarg Shah
HIGHLIGHT:      We propose a multi-agent variant of the classical multi-armed bandit problem, in which there are N agents and K arms, and pulling an arm generates a (possibly different) stochastic reward to each agent.

65, TITLE:      Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition, and Selective Transfer
http://arxiv.org/abs/2007.06918
AUTHORS:        Aswin Raghavan ; Jesse Hostetler ; Indranil Sur ; Abrar Rahman ; Ajay Divakaran
COMMENTS:       Accepted at the 4th Lifelong Machine Learning Workshop at the Thirty-seventh International Conference on Machine Learning (ICML) 2020
HIGHLIGHT:      We introduce the eigentask framework for lifelong learning.

66, TITLE:      AQD: Towards Accurate Quantized Object Detection
http://arxiv.org/abs/2007.06919
AUTHORS:        Jing Liu ; Bohan Zhuang ; Peng Chen ; Mingkui Tan ; Chunhua Shen
COMMENTS:       Code is available at https://github.com/blueardour/model-quantization
HIGHLIGHT:      In this paper, we demonstrate that the poor performance of the quantized network on object detection comes from the inaccurate batch statistics of batch normalization.

67, TITLE:      Questionnaire analysis to define the most suitable survey for port-noise investigation
http://arxiv.org/abs/2007.06915
AUTHORS:        Andrea Cerniglia ; Davide Chiarella ; Paola Cutugno ; Lucia Marconi ; Anna Magrini ; Gelsomina Di Feo ; Melissa Ferretti
COMMENTS:       8 pages, Proceedings of the 26th International Congress on Sound and Vibration. ISBN 978-1-9991810-0-0 ISSN 2329-3675
HIGHLIGHT:      The questionnaire will be optimized to be distributed in the TRIPLO project (TRansports and Innovative sustainable connections between Ports and LOgistic platforms).

68, TITLE:      An Interpretable Baseline for Time Series Classification Without Intensive Learning
http://arxiv.org/abs/2007.06682
AUTHORS:        Robert J. Ravier ; Mohammadreza Soltani ; Miguel Antunes Dias Alfaiate ; Denis Garagic ; Vahid Tarokh
COMMENTS:       21 pages, 9 tables, 4 figures
HIGHLIGHT:      In this paper, we propose an interpretable baseline based on representing each time series as a collection of probability distributions of extracted geometric features.

69, TITLE:      Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations
http://arxiv.org/abs/2007.06929
AUTHORS:        Hongyu Liu ; Bin Jiang ; Yibing Song ; Wei Huang ; Chao Yang
COMMENTS:       Accepted by ECCV2020(oral)
HIGHLIGHT:      In this paper, we propose a mutual encoder-decoder CNN for joint recovery of both.

70, TITLE:      Gradient Descent over Metagrammars for Syntax-Guided Synthesis
http://arxiv.org/abs/2007.06677
AUTHORS:        Nicolas Chan ; Elizabeth Polgreen ; Sanjit A. Seshia
COMMENTS:       4 pages, SYNT 2020
HIGHLIGHT:      In this work, we speculate this default grammar could be improved upon substantially. We build sets of rules, or metagrammars, for constructing grammars, and perform a gradient descent over these metagrammars aiming to find a metagrammar which solves more benchmarks and on average faster.

71, TITLE:      A Graph-based Interactive Reasoning for Human-Object Interaction Detection
http://arxiv.org/abs/2007.06925
AUTHORS:        Dongming Yang ; Yuexian Zou
COMMENTS:       Accepted by IJCAI 2020. SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence)
HIGHLIGHT:      In this paper, we present a novel graph-based interactive reasoning model called Interactive Graph (abbr.

72, TITLE:      Language, communication and society: a gender based linguistics analysis
http://arxiv.org/abs/2007.06908
AUTHORS:        P. Cutugno ; D. Chiarella ; R. Lucentini ; L. Marconi ; G. Morgavi
COMMENTS:       7 pages, Mladenov et al., Recent Advances in Communications - Proceedings of the 19th International Conference on Communications (part of 19th International Conference on Circuits, Systems, Communications and Computers 2015)
HIGHLIGHT:      The purpose of this study is to find evidence for supporting the hypothesis that language is the mirror of our thinking, our prejudices and cultural stereotypes.

73, TITLE:      Meta-rPPG: Remote Heart Rate Estimation Using a Transductive Meta-Learner
http://arxiv.org/abs/2007.06786
AUTHORS:        Eugene Lee ; Evan Chen ; Chen-Yi Lee
COMMENTS:       26 pages, 10 figures, accepted by European Conference on Computer Vision (ECCV) 2020
HIGHLIGHT:      Remote heart rate estimation is the measurement of heart rate without any physical contact with the subject and is accomplished using remote photoplethysmography (rPPG) in this work.

74, TITLE:      Vehicle Trajectory Prediction by Transfer Learning of Semi-Supervised Models
http://arxiv.org/abs/2007.06781
AUTHORS:        Nick Lamm ; Shashank Jaiprakash ; Malavika Srikanth ; Iddo Drori
HIGHLIGHT:      In this work we show that semi-supervised models for vehicle trajectory prediction significantly improve performance over supervised models on state-of-the-art real-world benchmarks.

75, TITLE:      Verification of ML Systems via Reparameterization
http://arxiv.org/abs/2007.06776
AUTHORS:        Jean-Baptiste Tristan ; Joseph Tassarotti ; Koundinya Vajjha ; Michael L. Wick ; Anindya Banerjee
HIGHLIGHT:      A growing body of research has developed machine learning algorithms with formal guarantees about performance, robustness, or fairness.

76, TITLE:      An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models
http://arxiv.org/abs/2007.06778
AUTHORS:        Lifu Tu ; Garima Lalwani ; Spandana Gella ; He He
COMMENTS:       Accepted to TACL 2020
HIGHLIGHT:      In the case of extreme minority, we propose to use multi-task learning (MTL) to improve generalization.

77, TITLE:      Calling Out Bluff: Attacking the Robustness of Automatic Scoring Systems with Simple Adversarial Testing
http://arxiv.org/abs/2007.06796
AUTHORS:        Yaman Kumar ; Mehar Bhatia ; Anubha Kabra ; Jessy Junyi Li ; Di Jin ; Rajiv Ratn Shah
HIGHLIGHT:      Inspired by common student behaviour during examinations, we propose a task agnostic adversarial evaluation scheme for AES systems to test their natural language understanding capabilities and overall robustness.

78, TITLE:      TCGM: An Information-Theoretic Framework for Semi-Supervised Multi-Modality Learning
http://arxiv.org/abs/2007.06793
AUTHORS:        Xinwei Sun ; Yilun Xu ; Peng Cao ; Yuqing Kong ; Lingjing Hu ; Shanghang Zhang ; Yizhou Wang
COMMENTS:       ECCV 2020 (oral)
HIGHLIGHT:      In this paper, we propose a novel information-theoretic approach, namely \textbf{T}otal \textbf{C}orrelation \textbf{G}ain \textbf{M}aximization (TCGM), for semi-supervised multi-modal learning, which is endowed with promising properties: (i) it can utilize effectively the information across different modalities of unlabeled data points to facilitate training classifiers of each modality (ii) it has theoretical guarantee to identify Bayesian classifiers, i.e., the ground truth posteriors of all modalities.

79, TITLE:      Can neural networks acquire a structural bias from raw linguistic data?
http://arxiv.org/abs/2007.06761
AUTHORS:        Alex Warstadt ; Samuel R. Bowman
COMMENTS:       To appear in Proceedings of 42nd Annual Meeting of the Cognitive Science Society
HIGHLIGHT:      We evaluate whether BERT, a widely used neural network for sentence processing, acquires an inductive bias towards forming structural generalizations through pretraining on raw data.

80, TITLE:      A Single Stream Network for Robust and Real-time RGB-D Salient Object Detection
http://arxiv.org/abs/2007.06811
AUTHORS:        Xiaoqi Zhao ; Lihe Zhang ; Youwei Pang ; Huchuan Lu ; Lei Zhang
COMMENTS:       Accepted in ECCV2020. Code: https://github.com/Xiaoqi-Zhao-DLUT/DANet-RGBD-Saliency
HIGHLIGHT:      In this work, we design a single stream network to directly use the depth map to guide early fusion and middle fusion between RGB and depth, which saves the feature encoder of the depth stream and achieves a lightweight and real-time model.

81, TITLE:      Synthesis in Uclid5
http://arxiv.org/abs/2007.06760
AUTHORS:        Federico Mora ; Kevin Cheang ; Elizabeth Polgreen ; Sanjit A. Seshia
HIGHLIGHT:      We describe an integration of program synthesis intoUclid5,a formal modelling and verification tool. We use the integration to generate 25 pro-gram synthesis benchmarks with simple, known solutions that are out of reach of current synthesis engines, and we release the benchmarks to the community.

82, TITLE:      RGB-D Salient Object Detection with Cross-Modality Modulation and Selection
http://arxiv.org/abs/2007.07051
AUTHORS:        Chongyi Li ; Runmin Cong ; Yongri Piao ; Qianqian Xu ; Chen Change Loy
COMMENTS:       ECCV2020
HIGHLIGHT:      We present an effective method to progressively integrate and refine the cross-modality complementarities for RGB-D salient object detection (SOD).

83, TITLE:      Unsupervised Human 3D Pose Representation with Viewpoint and Pose Disentanglement
http://arxiv.org/abs/2007.07053
AUTHORS:        Qiang Nie ; Ziwei Liu ; Yunhui Liu
COMMENTS:       To appear in ECCV 2020. Code and models are available at: https://github.com/NIEQiang001/unsupervised-human-pose.git
HIGHLIGHT:      In this work, we propose a novel Siamese denoising autoencoder to learn a 3D pose representation by disentangling the pose-dependent and view-dependent feature from the human skeleton data, in a fully unsupervised manner.

84, TITLE:      Personalized Face Modeling for Improved Face Reconstruction and Motion Retargeting
http://arxiv.org/abs/2007.06759
AUTHORS:        Bindita Chaudhuri ; Noranart Vesdapunt ; Linda Shapiro ; Baoyuan Wang
HIGHLIGHT:      We propose an end-to-end framework that jointly learns a personalized face model per user and per-frame facial motion parameters from a large corpus of in-the-wild videos of user expressions.

85, TITLE:      Consensus Halving for Sets of Items
http://arxiv.org/abs/2007.06754
AUTHORS:        Paul W. Goldberg ; Alexandros Hollender ; Ayumi Igarashi ; Pasin Manurangsi ; Warut Suksompong
HIGHLIGHT:      In this paper, we study consensus halving in a natural setting where the resource consists of a set of items without a linear ordering.

86, TITLE:      JNR: Joint-based Neural Rig Representation for Compact 3D Face Modeling
http://arxiv.org/abs/2007.06755
AUTHORS:        Noranart Vesdapunt ; Mitch Rundle ; HsiangTao Wu ; Baoyuan Wang
HIGHLIGHT:      In this paper, we introduce a novel approach to learn a 3D face model using a joint-based face rig and a neural skinning network.

87, TITLE:      Towards Realistic 3D Embedding via View Alignment
http://arxiv.org/abs/2007.07066
AUTHORS:        Fangneng Zhan ; Shijian Lu ; Changgong Zhang ; Feiying Ma ; Xuansong Xie
COMMENTS:       12 pages, 7 figures
HIGHLIGHT:      This paper presents an innovative View Alignment GAN (VA-GAN) that composes new images by embedding 3D models into 2D background images realistically and automatically.

88, TITLE:      Semi-supervised Learning with a Teacher-student Network for Generalized Attribute Prediction
http://arxiv.org/abs/2007.06769
AUTHORS:        Minchul Shin
COMMENTS:       14 pages, Accepted to ECCV 2020
HIGHLIGHT:      With that in mind, we propose a multi-teacher-single-student (MTSS) approach inspired by the multi-task learning and the distillation of semi-supervised learning.

89, TITLE:      Patch-wise Attack for Fooling Deep Neural Network
http://arxiv.org/abs/2007.06765
AUTHORS:        Lianli Gao ; Qilong Zhang ; Jingkuan Song ; Xianglong Liu ; Heng Tao Shen
COMMENTS:       21 pages, 3 figures
HIGHLIGHT:      Motivated by this, we propose a patch-wise iterative algorithm -- a black-box attack towards mainstream normally trained and defense models, which differs from the existing attack methods manipulating pixel-wise noise.

90, TITLE:      Algorithmic applications of the corestriction of central simple algebras
http://arxiv.org/abs/2007.06981
AUTHORS:        Tímea Csahók ; Péter Kutas ; Gergely Zábrádi
COMMENTS:       22 pages
HIGHLIGHT:      We propose efficient algorithms for finding isomorphisms between quaternion algebras over $L$.

91, TITLE:      DeepMSRF: A novel Deep Multimodal Speaker Recognition framework with Feature selection
http://arxiv.org/abs/2007.06809
AUTHORS:        Ehsan Asali ; Farzan Shenavarmasouleh ; Farid Ghareh Mohammadi ; Prasanth Sengadu Suresh ; Hamid R. Arabnia
COMMENTS:       The 24th International Conference on Image Processing, Computer Vision, & Pattern Recognition (IPCV'20: July 27-30, 2020, USA)
HIGHLIGHT:      In this paper, we address this problem from a different perspective and propose an unprecedented multimodality data fusion framework called DeepMSRF, Deep Multimodal Speaker Recognition with Feature selection.

92, TITLE:      Quantum Software Engineering: Landscapes and Horizons
http://arxiv.org/abs/2007.07047
AUTHORS:        Jianjun Zhao
HIGHLIGHT:      This paper defines the term "quantum software engineering" and introduces a quantum software life cycle. Based on these, the paper provides a comprehensive survey of the current state of the art in the field and presents the challenges and opportunities that we face.

93, TITLE:      MFRNet: A New CNN Architecture for Post-Processing and In-loop Filtering
http://arxiv.org/abs/2007.07099
AUTHORS:        Di Ma ; Fan Zhang ; David R. Bull
HIGHLIGHT:      In this paper, we propose a novel convolutional neural network (CNN) architecture, MFRNet, for post-processing (PP) and in-loop filtering (ILF) in the context of video compression.

94, TITLE:      Lifelong Policy Gradient Learning of Factored Policies for Faster Training Without Forgetting
http://arxiv.org/abs/2007.07011
AUTHORS:        Jorge A. Mendez ; Boyu Wang ; Eric Eaton
HIGHLIGHT:      We provide a novel method for lifelong policy gradient learning that trains lifelong function approximators directly via policy gradients, allowing the agent to benefit from accumulated knowledge throughout the entire training process.

95, TITLE:      Pose2RGBD. Generating Depth and RGB images from absolute positions
http://arxiv.org/abs/2007.07013
AUTHORS:        Mihai Cristian Pîrvu
HIGHLIGHT:      We propose a method at the intersection of Computer Vision and Computer Graphics fields, which automatically generates RGBD images using neural networks, based on previously seen and synchronized video, depth and pose signals. We introduce two new datasets, one based on synthetic data with full ground truth information, while the other one being recorded from a drone flight in an university campus, using only video and GPS signals.

96, TITLE:      Pasadena: Perceptually Aware and Stealthy Adversarial Denoise Attack
http://arxiv.org/abs/2007.07097
AUTHORS:        Yupeng Cheng ; Qing Guo ; Felix Juefei-Xu ; Xiaofei Xie ; Shang-Wei Lin ; Weisi Lin ; Wei Feng ; Yang Liu
HIGHLIGHT:      In this paper, we investigate a new task, adversarial denoise attack, that stealthily embeds attacks inside the image denoising module.

97, TITLE:      A Normative approach to Attest Digital Discrimination
http://arxiv.org/abs/2007.07092
AUTHORS:        Natalia Criado ; Xavier Ferrer ; Jose M. Such
COMMENTS:       Author's copy of the manuscript accepted in the Advancing Towards the SDGS Artificial Intelligence for a Fair, Just and Equitable World Workshop of the 24th European Conference on Artificial Intelligence (ECAI'20)
HIGHLIGHT:      In this paper, we use norms as an abstraction to represent different situations that may lead to digital discrimination.

98, TITLE:      Video Object Segmentation with Episodic Graph Memory Networks
http://arxiv.org/abs/2007.07020
AUTHORS:        Xinkai Lu ; Wenguan Wang ; Martin Danelljan ; Tianfei Zhou ; Jianbing Shen ; Luc Van Gool
COMMENTS:       ECCV2020 Spotlight; website: https://github.com/carrierlxk/GraphMemVOS
HIGHLIGHT:      In this work, a novel graph memory network is developed to address the novel idea of ``learning to update the segmentation model''.

99, TITLE:      Unsupervised Multi-Target Domain Adaptation Through Knowledge Distillation
http://arxiv.org/abs/2007.07077
AUTHORS:        Le Thanh Nguyen-Meidine ; Madhu Kiran ; Jose Dolz ; Eric Granger ; Atif Bela ; Louis-Antoine Blais-Morin
COMMENTS:       Submitted to WACV-2021
HIGHLIGHT:      In this paper, we propose a new unsupervised MTDA approach to train a common CNN that can generalize across multiple target domains.

100, TITLE:     UDBNET: Unsupervised Document Binarization Network via Adversarial Game
http://arxiv.org/abs/2007.07075
AUTHORS:        Amandeep Kumar ; Shuvozit Ghose ; Pinaki Nath Chowdhury ; Partha Pratim Roy ; Umapada Pal
COMMENTS:       Submitted to ICPR 2020
HIGHLIGHT:      In this paper, we present a novel approach towards document image binarization by introducing three-player min-max adversarial game.



==========Updates to Previous Papers==========
1, TITLE:       Towards Linear Time Neural Machine Translation with Capsule Networks
http://arxiv.org/abs/1811.00287
AUTHORS:        Mingxuan Wang ; Jun Xie ; Zhixing Tan ; Jinsong Su ; Deyi Xiong ; Lei Li
COMMENTS:       Accepted as EMNLP2019
HIGHLIGHT:      To the best of our knowledge, this is the first work that capsule networks have been empirically investigated for sequence to sequence problems.

2, TITLE:       Target-Independent Domain Adaptation for WBC Classification using Generative Latent Search
http://arxiv.org/abs/2005.05432
AUTHORS:        Prashant Pandey ; Prathosh AP ; Vinay Kyatham ; Deepak Mishra ; Tathagato Rai Dastidar
COMMENTS:       IEEE TMI 2020
HIGHLIGHT:      In this paper, we propose a method for UDA that is devoid of the need for target data.

3, TITLE:       Emergent Multi-Agent Communication in the Deep Learning Era
http://arxiv.org/abs/2006.02419
AUTHORS:        Angeliki Lazaridou ; Marco Baroni
COMMENTS:       Added some more references and discussion
HIGHLIGHT:      Emergent Multi-Agent Communication in the Deep Learning Era

4, TITLE:       Hierarchical Kinematic Human Mesh Recovery
http://arxiv.org/abs/2003.04232
AUTHORS:        Georgios Georgakis ; Ren Li ; Srikrishna Karanam ; Terrence Chen ; Jana Kosecka ; Ziyan Wu
COMMENTS:       17 pages, 8 figures, 5 tables, ECCV 2020
HIGHLIGHT:      In this work, we address this gap by proposing a new technique for regression of human parametric model that is explicitly informed by the known hierarchical structure, including joint interdependencies of the model.

5, TITLE:       Provably Efficient Exploration for RL with Unsupervised Learning
http://arxiv.org/abs/2003.06898
AUTHORS:        Fei Feng ; Ruosong Wang ; Wotao Yin ; Simon S. Du ; Lin F. Yang
HIGHLIGHT:      We present a general algorithmic framework that is built upon two components: an unsupervised learning algorithm and a no-regret tabular RL algorithm.

6, TITLE:       Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models
http://arxiv.org/abs/2007.05791
AUTHORS:        Yue Liu ; Hossein Azizpour ; Fredrik Strand ; Kevin Smith
COMMENTS:       Medical Image Computing and Computer Assisted Interventions 2020
HIGHLIGHT:      With this in mind, we trained networks using three different criteria to select the positive training data (i.e. images from patients that will develop cancer): an inherent risk model trained on images with no visible signs of cancer, a cancer signs model trained on images containing cancer or early signs of cancer, and a conflated model trained on all images from patients with a cancer diagnosis.

7, TITLE:       Anomaly Detection by One Class Latent Regularized Networks
http://arxiv.org/abs/2002.01607
AUTHORS:        Chengwei Chen ; Pan Chen ; Haichuan Song ; Yiqing Tao ; Yuan Xie ; Shouhong Ding ; Lizhuang Ma
HIGHLIGHT:      Given a wide range of images belonging to the normal class, emerging from some distribution, the objective of this task is to construct the model to detect out-of-distribution images belonging to abnormal instances.

8, TITLE:       Spatio-Temporal Event Segmentation and Localization for Wildlife Extended Videos
http://arxiv.org/abs/2005.02463
AUTHORS:        Ramy Mounir ; Roman Gula ; Jörn Theuerkauf ; Sudeep Sarkar
HIGHLIGHT:      We present a self-supervised perceptual prediction framework capable of temporal event segmentation by building stable representations of objects over time and demonstrate it on long videos, spanning several days. We will make the dataset, which is the first of its kind, and the code available to the research community.

9, TITLE:       Learning Variable Impedance Control for Contact Sensitive Tasks
http://arxiv.org/abs/1907.07500
AUTHORS:        Miroslav Bogdanovic ; Majid Khadiv ; Ludovic Righetti
HIGHLIGHT:      In this paper, we investigate how the choice of action space can give robust performance in presence of contact uncertainties.

10, TITLE:      Abstract Universal Approximation for Neural Networks
http://arxiv.org/abs/2007.06093
AUTHORS:        Zi Wang ; Aws Albarghouthi ; Somesh Jha
HIGHLIGHT:      We present a theoretical result that demonstrates the power of numerical domains, namely, the simple interval domain, for analysis of neural networks.

11, TITLE:      Phylogenetic signal in phonotactics
http://arxiv.org/abs/2002.00527
AUTHORS:        Jayden L. Macklin-Cordes ; Claire Bowern ; Erich R. Round
COMMENTS:       Main text: 32 pages, 17 figures, 1 table. Supplementary Information: 17 pages, 1 figure. Code and data available at http://doi.org/10.5281/zenodo.3936353. This article is in review but not yet accepted for publication in a journal
HIGHLIGHT:      Nevertheless, we detect phylogenetic signal in all datasets.

12, TITLE:      Near-Optimal Reinforcement Learning with Self-Play
http://arxiv.org/abs/2006.12007
AUTHORS:        Yu Bai ; Chi Jin ; Tiancheng Yu
HIGHLIGHT:      This paper closes this gap for the first time: we propose an optimistic variant of the \emph{Nash Q-learning} algorithm with sample complexity $\tilde{\mathcal{O}}(SAB)$, and a new \emph{Nash V-learning} algorithm with sample complexity $\tilde{\mathcal{O}}(S(A+B))$.

13, TITLE:      Bridging Knowledge Graphs to Generate Scene Graphs
http://arxiv.org/abs/2001.02314
AUTHORS:        Alireza Zareian ; Svebor Karaman ; Shih-Fu Chang
COMMENTS:       To be presented at ECCV 2020
HIGHLIGHT:      In this paper, we present a unified formulation of these two constructs, where a scene graph is seen as an image-conditioned instantiation of a commonsense knowledge graph.

14, TITLE:      A Comparative Study on Early Detection of COVID-19 from Chest X-Ray Images
http://arxiv.org/abs/2006.05332
AUTHORS:        Mete Ahishali ; Aysen Degerli ; Mehmet Yamac ; Serkan Kiranyaz ; Muhammad E. H. Chowdhury ; Khalid Hameed ; Tahir Hamid ; Rashid Mazhar ; Moncef Gabbouj
COMMENTS:       10 pages
HIGHLIGHT:      In this study, our first aim is to evaluate the ability of recent state-of-the-art Machine Learning techniques for the early detection of COVID-19 from chest X-ray images. Finally, this study introduces a new benchmark dataset called Early-QaTa-COV19, which consists of 175 early-stage COVID-19 Pneumonia samples (very limited or no infection signs) labelled by the medical doctors and 1579 samples for control (normal) class.

15, TITLE:      Training Object Detectors from Few Weakly-Labeled and Many Unlabeled Images
http://arxiv.org/abs/1912.00384
AUTHORS:        Zhaohui Yang ; Miaojing Shi ; Chao Xu ; Vittorio Ferrari ; Yannis Avrithis
HIGHLIGHT:      In this work, we study the problem of training an object detector from one or few images with image-level labels and a larger set of completely unlabeled images.

16, TITLE:      OVC-Net: Object-Oriented Video Captioning with Temporal Graph and Detail Enhancement
http://arxiv.org/abs/2003.03715
AUTHORS:        Fangyi Zhu ; Jenq-Neng Hwang ; Zhanyu Ma ; Guang Chen ; Jun Guo
HIGHLIGHT:      In this paper, we propose a novel task to understand the videos in object-level, named object-oriented video captioning. Thereafter, we construct a new dataset, providing consistent object-sentence pairs, to facilitate effective cross-modal learning.

17, TITLE:      Learning Combinatorial Optimization on Graphs: A Survey with Applications to Networking
http://arxiv.org/abs/2005.11081
AUTHORS:        Natalia Vesselinova ; Rebecca Steinert ; Daniel F. Perez-Ramirez ; Magnus Boman
COMMENTS:       29 pages, 1 figure, open access journal publication
HIGHLIGHT:      Relevant developments in machine learning research on graphs are surveyed for this purpose.

18, TITLE:      Parameterized Approximation Schemes for Steiner Trees with Small Number of Steiner Vertices
http://arxiv.org/abs/1710.00668
AUTHORS:        Pavel Dvořák ; Andreas Emil Feldmann ; Dušan Knop ; Tomáš Masařík ; Tomáš Toufar ; Pavel Veselý
COMMENTS:       23 pages, 6 figures An extended abstract appeared in proceedings of STACS 2018
HIGHLIGHT:      We study the Steiner Tree problem, in which a set of terminal vertices needs to be connected in the cheapest possible way in an edge-weighted graph.

19, TITLE:      Aligning Superhuman AI with Human Behavior: Chess as a Model System
http://arxiv.org/abs/2006.01855
AUTHORS:        Reid McIlroy-Young ; Siddhartha Sen ; Jon Kleinberg ; Ashton Anderson
COMMENTS:       11 pages, 11 figure, Proceedings of the 25th ACM SIGKDD international conference on Knowledge discovery and data mining, Virtual 2020
HIGHLIGHT:      We develop and introduce Maia, a customized version of Alpha-Zero trained on human chess games, that predicts human moves at a much higher accuracy than existing engines, and can achieve maximum accuracy when predicting decisions made by players at a specific skill level in a tuneable way.

20, TITLE:      Temporal Self-Ensembling Teacher for Semi-Supervised Object Detection
http://arxiv.org/abs/2007.06144
AUTHORS:        Cong Chen ; Shouyang Dong ; Ye Tian ; Kunlin Cao ; Li Liu ; Yuanhao Guo
COMMENTS:       11 papges, 3 figures, preprint for submission
HIGHLIGHT:      To solve these problems, we propose the Temporal Self-Ensembling Teacher (TSE-T) model on top of the KD framework.

21, TITLE:      Computing the k Densest Subgraphs of a Graph
http://arxiv.org/abs/2002.07695
AUTHORS:        Riccardo Dondi ; Danny Hermelin
HIGHLIGHT:      In this paper we hope to remedy this situation by analyzing three natural variants of the k densest subgraphs problem.

22, TITLE:      Effects for Efficiency: Asymptotic Speedup with First-Class Control
http://arxiv.org/abs/2007.00605
AUTHORS:        Daniel Hillerström ; Sam Lindley ; John Longley
HIGHLIGHT:      We consider the generic count problem using a pure PCF-like base language $\lambda_b$ and its extension with effect handlers $\lambda_h$.

23, TITLE:      BERTology Meets Biology: Interpreting Attention in Protein Language Models
http://arxiv.org/abs/2006.15222
AUTHORS:        Jesse Vig ; Ali Madani ; Lav R. Varshney ; Caiming Xiong ; Richard Socher ; Nazneen Fatema Rajani
HIGHLIGHT:      Through the lens of attention, we analyze the inner workings of the Transformer and explore how the model discerns structural and functional properties of proteins.

24, TITLE:      Are Natural Language Inference Models IMPPRESsive? Learning IMPlicature and PRESupposition
http://arxiv.org/abs/2004.03066
AUTHORS:        Paloma Jeretic ; Alex Warstadt ; Suvrat Bhooshan ; Adina Williams
COMMENTS:       to appear in Proceedings of ACL 2020
HIGHLIGHT:      We create an IMPlicature and PRESupposition diagnostic dataset (IMPPRES), consisting of >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types.

25, TITLE:      Symmetric Dilated Convolution for Surgical Gesture Recognition
http://arxiv.org/abs/2007.06373
AUTHORS:        Jinglu Zhang ; Yinyu Nie ; Yao Lyu ; Hailin Li ; Jian Chang ; Xiaosong Yang ; Jian Jun Zhang
COMMENTS:       Accepted to MICCAI 2020
HIGHLIGHT:      To tackle these challenges, we propose a novel temporal convolutional architecture to automatically detect and segment surgical gestures with corresponding boundaries only using RGB videos.

26, TITLE:      Neural Networks on Random Graphs
http://arxiv.org/abs/2002.08104
AUTHORS:        Romuald A. Janik ; Aleksandra Nowak
COMMENTS:       Corrected typos, extended supplementary materials, rephrased section 5.6
HIGHLIGHT:      Apart from the classical random graph families including random, scale-free and small world graphs, we introduced a novel and flexible algorithm for directly generating random directed acyclic graphs (DAG) and studied a class of graphs derived from functional resting state fMRI networks.

27, TITLE:      Multiple Sound Sources Localization from Coarse to Fine
http://arxiv.org/abs/2007.06355
AUTHORS:        Rui Qian ; Di Hu ; Heinrich Dinkel ; Mengyue Wu ; Ning Xu ; Weiyao Lin
COMMENTS:       to appear in ECCV 2020
HIGHLIGHT:      To solve this problem, we develop a two-stage audiovisual learning framework that disentangles audio and visual representations of different categories from complex scenes, then performs cross-modal feature alignment in a coarse-to-fine manner.

28, TITLE:      Reference-guided Face Component Editing
http://arxiv.org/abs/2006.02051
AUTHORS:        Qiyao Deng ; Jie Cao ; Yunfan Liu ; Zhenhua Chai ; Qi Li ; Zhenan Sun
HIGHLIGHT:      To break the limitations (e.g. shape, mask or sketch) of the existing methods, we propose a novel framework termed r-FACE (Reference-guided FAce Component Editing) for diverse and controllable face component editing with geometric changes.

29, TITLE:      SSM-Net for Plants Disease Identification in Low Data Regime
http://arxiv.org/abs/2005.13140
AUTHORS:        Shruti Jadon
COMMENTS:       5 pages, 7 Figures
HIGHLIGHT:      In this paper, we propose a new metrics-based few-shot learning SSM net architecture, which consists of stacked siamese and matching network components to solve the problem of disease detection in low data regimes.

30, TITLE:      An argument in favor of strong scaling for deep neural networks with small datasets
http://arxiv.org/abs/1807.09161
AUTHORS:        Renato L. de F. Cunha ; Eduardo R. Rodrigues ; Matheus Palhares Viana ; Dario Augusto Borges Oliveira
COMMENTS:       8 pages, 5 figures, Presented at HPML 2018 - http://hpml2018.github.io/
HIGHLIGHT:      We evaluated our approach with up to 32 GPUs and show that weak scaling not only does not have the same accuracy as the sequential model, it also fails to converge most of time.

31, TITLE:      Learning Object Permanence from Video
http://arxiv.org/abs/2003.10469
AUTHORS:        Aviv Shamsian ; Ofri Kleinfeld ; Amir Globerson ; Gal Chechik
COMMENTS:       16th european conference on computer vision (ECCV 2020)
HIGHLIGHT:      Here we introduce the setup of learning Object Permanence from data.

32, TITLE:      Wavelet Integrated CNNs for Noise-Robust Image Classification
http://arxiv.org/abs/2005.03337
AUTHORS:        Qiufu Li ; Linlin Shen ; Sheng Guo ; Zhihui Lai
COMMENTS:       CVPR accepted paper
HIGHLIGHT:      We present general DWT and Inverse DWT (IDWT) layers applicable to various wavelets like Haar, Daubechies, and Cohen, etc., and design wavelet integrated CNNs (WaveCNets) using these layers for image classification.

33, TITLE:      MvMM-RegNet: A new image registration framework based on multivariate mixture model and neural network estimation
http://arxiv.org/abs/2006.15573
AUTHORS:        Xinzhe Luo ; Xiahai Zhuang
COMMENTS:       Accepted for publication at MICCAI 2020; Code is available from https://github.com/xzluo97/MvMM-RegNet
HIGHLIGHT:      In this paper, we propose a new image registration framework based on multivariate mixture model (MvMM) and neural network estimation.

34, TITLE:      PolyLaneNet: Lane Estimation via Deep Polynomial Regression
http://arxiv.org/abs/2004.10924
AUTHORS:        Lucas Tabelini ; Rodrigo Berriel ; Thiago M. Paixão ; Claudine Badue ; Alberto F. De Souza ; Thiago Oliveira-Santos
COMMENTS:       Accepted to ICPR 2020
HIGHLIGHT:      In this work, we present a novel method for lane detection that uses as input an image from a forward-looking camera mounted in the vehicle and outputs polynomials representing each lane marking in the image, via deep polynomial regression.

35, TITLE:      Neural Knowledge Extraction From Cloud Service Incidents
http://arxiv.org/abs/2007.05505
AUTHORS:        Manish Shetty ; Chetan Bansal ; Sumit Kumar ; Nikitha Rao ; Nachiappan Nagappan ; Thomas Zimmermann
HIGHLIGHT:      In this work, we address the fundamental problem of structured knowledge extraction from service incidents.

36, TITLE:      Relational Deep Feature Learning for Heterogeneous Face Recognition
http://arxiv.org/abs/2003.00697
AUTHORS:        MyeongAh Cho ; Taeoh Kim ; Ig-Jae Kim ; Kyungjae Lee ; Sangyoun Lee
HIGHLIGHT:      With this motivation, we propose a graph-structured module called Relational Graph Module (RGM) that extracts global relational information in addition to general facial features.

37, TITLE:      Learning Bijective Feature Maps for Linear ICA
http://arxiv.org/abs/2002.07766
AUTHORS:        Alexander Camuto ; Matthew Willetts ; Brooks Paige ; Chris Holmes ; Stephen Roberts
COMMENTS:       10 pages
HIGHLIGHT:      Given the complexities of training such a hybrid model, we introduce novel theory that constrains linear ICA to lie close to the manifold of decorrelating matrices, the Stiefel manifold.

38, TITLE:      MeDaS: An open-source platform as service to help break the walls between medicine and informatics
http://arxiv.org/abs/2007.06013
AUTHORS:        Liang Zhang ; Johann Li ; Ping Li ; Xiaoyuan Lu ; Peiyi Shen ; Guangming Zhu ; Syed Afaq Shah ; Mohammed Bennarmoun ; Kun Qian ; Björn W. Schuller
COMMENTS:       layout error fixed
HIGHLIGHT:      Based on a series of toolkits and utilities from the idea of RINV (Rapid Implementation aNd Verification), our proposed MeDaS platform can implement pre-processing, post-processing, augmentation, visualization, and other phases needed in medical image analysis.

39, TITLE:      DeepVS: An Efficient and Generic Approach for Source Code Modeling Usage
http://arxiv.org/abs/1910.06500
AUTHORS:        Yasir Hussain ; Zhiqiu Huang ; Yu Zhou ; Senzhang Wang
HIGHLIGHT:      This letter presents \textit{DeepVS}, an end-to-end deep neural code completion tool that learns from existing codebases by exploiting the bidirectional Gated Recurrent Unit (BiGRU) neural net.

40, TITLE:      HuggingFace's Transformers: State-of-the-art Natural Language Processing
http://arxiv.org/abs/1910.03771
AUTHORS:        Thomas Wolf ; Lysandre Debut ; Victor Sanh ; Julien Chaumond ; Clement Delangue ; Anthony Moi ; Pierric Cistac ; Tim Rault ; Rémi Louf ; Morgan Funtowicz ; Joe Davison ; Sam Shleifer ; Patrick von Platen ; Clara Ma ; Yacine Jernite ; Julien Plu ; Canwen Xu ; Teven Le Scao ; Sylvain Gugger ; Mariama Drame ; Quentin Lhoest ; Alexander M. Rush
COMMENTS:       8 pages, 4 figures, more details at https://github.com/huggingface/transformers
HIGHLIGHT:      HuggingFace's Transformers: State-of-the-art Natural Language Processing

41, TITLE:      Ultra-Low-Power FDSOI Neural Circuits for Extreme-Edge Neuromorphic Intelligence
http://arxiv.org/abs/2006.14270
AUTHORS:        Arianna Rubino ; Can Livanelioglu ; Ning Qiao ; Melika Payvand ; Giacomo Indiveri
COMMENTS:       11 pages, 9 figures, TCAS submission
HIGHLIGHT:      To reduce power consumption even further, we present a set of mixed-signal analog/digital circuits that exploit the features of advanced Fully-Depleted Silicon on Insulator (FDSOI) integration processes.

42, TITLE:      How Does an Approximate Model Help in Reinforcement Learning?
http://arxiv.org/abs/1912.02986
AUTHORS:        Fei Feng ; Wotao Yin ; Lin F. Yang
HIGHLIGHT:      In this paper, we study the sample complexity of RL while an approximate model of the environment is provided.

43, TITLE:      Progressive Skeletonization: Trimming more fat from a network at initialization
http://arxiv.org/abs/2006.09081
AUTHORS:        Pau de Jorge ; Amartya Sanyal ; Harkirat S. Behl ; Philip H. S. Torr ; Gregory Rogez ; Puneet K. Dokania
HIGHLIGHT:      To this end, we propose to find a skeletonized network with maximum foresight connection sensitivity (FORCE).

44, TITLE:      Learning Reasoning Strategies in End-to-End Differentiable Proving
http://arxiv.org/abs/2007.06477
AUTHORS:        Pasquale Minervini ; Sebastian Riedel ; Pontus Stenetorp ; Edward Grefenstette ; Tim Rocktäschel
COMMENTS:       Proceedings of the 37th International Conference on Machine Learning (ICML 2020)
HIGHLIGHT:      We present Conditional Theorem Provers (CTPs), an extension to NTPs that learns an optimal rule selection strategy via gradient-based optimisation.

45, TITLE:      Towards Best Practice in Explaining Neural Network Decisions with LRP
http://arxiv.org/abs/1910.09840
AUTHORS:        Maximilian Kohlbrenner ; Alexander Bauer ; Shinichi Nakajima ; Alexander Binder ; Wojciech Samek ; Sebastian Lapuschkin
COMMENTS:       7 pages, 4 figures, 1 table. fixed table row compared to v2. Presented virtually at IJCNN 2020
HIGHLIGHT:      In this paper, we focus on a popular and widely used method of XAI, the Layer-wise Relevance Propagation (LRP).

46, TITLE:      Hierarchical Dynamic Filtering Network for RGB-D Salient Object Detection
http://arxiv.org/abs/2007.06227
AUTHORS:        Youwei Pang ; Lihe Zhang ; Xiaoqi Zhao ; Huchuan Lu
COMMENTS:       The work has been accepted by ECCV 2020. In the current version, the "supplementary material" in the text has been replaced by "appendix"
HIGHLIGHT:      In this paper, we explore these issues from a new perspective.

47, TITLE:      Semantically Plausible and Diverse 3D Human Motion Prediction
http://arxiv.org/abs/1912.08521
AUTHORS:        Sadegh Aliakbarian ; Fatemeh Sadat Saleh ; Mathieu Salzmann ; Lars Petersson ; Stephen Gould
HIGHLIGHT:      In this paper, we address both of these problems by developing a new variational framework that accounts for both diversity and semantic of the generated future motion.

48, TITLE:      Graph Neural Networks for Decentralized Multi-Robot Path Planning
http://arxiv.org/abs/1912.06095
AUTHORS:        Qingbiao Li ; Fernando Gama ; Alejandro Ribeiro ; Amanda Prorok
COMMENTS:       This paper has been accepted in the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020. For the simulation demo, see this https URL "https://youtu.be/AGDk2RozpMQ"
HIGHLIGHT:      To side-step these issues and move beyond hand-crafted heuristics, we propose a combined model that automatically synthesizes local communication and decision-making policies for robots navigating in constrained workspaces.

49, TITLE:      Making sense of sensory input
http://arxiv.org/abs/1910.02227
AUTHORS:        Richard Evans ; Jose Hernandez-Orallo ; Johannes Welbl ; Pushmeet Kohli ; Marek Sergot
HIGHLIGHT:      Our second contribution is a computer implementation, the Apperception Engine, that was designed to satisfy the above requirements.

50, TITLE:      Approximation Algorithms for Multi-Robot Patrol-Scheduling with Min-Max Latency
http://arxiv.org/abs/2005.02530
AUTHORS:        Peyman Afshani ; Mark De Berg ; Kevin Buchin ; Jie Gao ; Maarten Loffler ; Amir Nayyeri ; Benjamin Raichel ; Rik Sarkar ; Haotian Wang ; Hao-Tsung Yang
COMMENTS:       Proceedings of the 14th International Workshop on the Algorithmic Foundations of Robotics (WAFR 20)
HIGHLIGHT:      We present a polynomial-time algorithm with an approximation factor of $O(k^2 \log \frac{w_{\max}}{w_{\min}})$ to the optimal solution, where $w_{\max}$ and $w_{\min}$ are the maximum and minimum weight of the sites respectively.

51, TITLE:      Deep Transfer Learning for Source Code Modeling
http://arxiv.org/abs/1910.05493
AUTHORS:        Yasir Hussain ; Zhiqiu Huang ; Yu Zhou ; Senzhang Wang
HIGHLIGHT:      In this work, we propose a transfer learning-based approach that significantly improves the performance of deep learning-based source code models.

52, TITLE:      Estimate of the Neural Network Dimension Using Algebraic Topology and Lie Theory
http://arxiv.org/abs/2004.02881
AUTHORS:        Luciano Melodia ; Richard Lenz
COMMENTS:       The title of this article was formerly "Parameterization of Neural Networks with Connected Abelian Lie Groups as Data Manifold"
HIGHLIGHT:      In this paper, we present an approach to determine the smallest possible number of neurons in a layer of a neural network in such a way that the topology of the input space can be learned sufficiently well.

53, TITLE:      Continual Adaptation for Deep Stereo
http://arxiv.org/abs/2007.05233
AUTHORS:        Matteo Poggi ; Alessio Tonioni ; Fabio Tosi ; Stefano Mattoccia ; Luigi Di Stefano
COMMENTS:       Extended version of CVPR 2019 paper "Real-time self-adaptive deep stereo"
HIGHLIGHT:      Purposely, we propose a continual adaptation paradigm for deep stereo networks designed to deal with challenging and ever-changing environments.

54, TITLE:      Deep Active Learning via Open Set Recognition
http://arxiv.org/abs/2007.02196
AUTHORS:        Jaya Krishna Mandivarapu ; Blake Camp ; Rolando Estrada
COMMENTS:       Withdrawn to address fundamental concerns with the text
HIGHLIGHT:      Here, we formulate active learning as an open-set recognition problem.

55, TITLE:      BézierSketch: A generative model for scalable vector sketches
http://arxiv.org/abs/2007.02190
AUTHORS:        Ayan Das ; Yongxin Yang ; Timothy Hospedales ; Tao Xiang ; Yi-Zhe Song
COMMENTS:       Accepted as poster at ECCV 2020
HIGHLIGHT:      In this paper we present B\'ezierSketch, a novel generative model for fully vector sketches that are automatically scalable and high-resolution.

56, TITLE:      CodeGRU: Context-aware Deep Learning with Gated Recurrent Unit for Source Code Modeling
http://arxiv.org/abs/1903.00884
AUTHORS:        Yasir Hussain ; Zhiqiu Huang ; Yu Zhou ; Senzhang Wang
HIGHLIGHT:      We introduce a novel approach which can capture the source code context by leveraging the source code token types.

57, TITLE:      Unsupervised Abnormality Detection Using Heterogeneous Autonomous Systems
http://arxiv.org/abs/2006.03733
AUTHORS:        Sayeed Shafayet Chowdhury ; Kazi Mejbaul Islam ; Rouhan Noor
HIGHLIGHT:      To that effect, in this paper, a heterogeneous system is proposed which estimates the degree of abnormality of an unmanned surveillance drone, analyzing real-time image and IMU (Inertial Measurement Unit) sensor data in an unsupervised manner.

58, TITLE:      GIQA: Generated Image Quality Assessment
http://arxiv.org/abs/2003.08932
AUTHORS:        Shuyang Gu ; Jianmin Bao ; Dong Chen ; Fang Wen
COMMENTS:       ECCV2020
HIGHLIGHT:      We introduce three GIQA algorithms from two perspectives: learning-based and data-based.

59, TITLE:      Towards Real-Time Multi-Object Tracking
http://arxiv.org/abs/1909.12605
AUTHORS:        Zhongdao Wang ; Liang Zheng ; Yixuan Liu ; Yali Li ; Shengjin Wang
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this paper, we propose an MOT system that allows target detection and appearance embedding to be learned in a shared model.

60, TITLE:      Structure-Invariant Testing for Machine Translation
http://arxiv.org/abs/1907.08710
AUTHORS:        Pinjia He ; Clara Meister ; Zhendong Su
COMMENTS:       Accepted at ICSE 2020
HIGHLIGHT:      To tackle this challenge, we introduce structure-invariant testing (SIT), a novel metamorphic testing approach for validating machine translation software.

61, TITLE:      Point-Set Anchors for Object Detection, Instance Segmentation and Pose Estimation
http://arxiv.org/abs/2007.02846
AUTHORS:        Fangyun Wei ; Xiao Sun ; Hongyang Li ; Jingdong Wang ; Stephen Lin
COMMENTS:       To appear in ECCV 2020
HIGHLIGHT:      To facilitate inference, we propose to instead perform regression from a set of points placed at more advantageous positions.

62, TITLE:      Spatial-Adaptive Network for Single Image Denoising
http://arxiv.org/abs/2001.10291
AUTHORS:        Meng Chang ; Qi Li ; Huajun Feng ; Zhihai Xu
HIGHLIGHT:      In this paper, we propose a novel spatial-adaptive denoising network (SADNet) for efficient single image blind noise removal.

63, TITLE:      Visually Guided Sound Source Separation using Cascaded Opponent Filter Network
http://arxiv.org/abs/2006.03028
AUTHORS:        Lingyu Zhu ; Esa Rahtu
COMMENTS:       main paper 14 pages, ref 3 pages, and supp 7 pages. Revised argument in section 3 and 4
HIGHLIGHT:      The objective of this paper is to recover the original component signals from a mixture audio with the aid of visual cues of the sound sources.

64, TITLE:      Scalable Planning with Deep Neural Network Learned Transition Models
http://arxiv.org/abs/1904.02873
AUTHORS:        Ga Wu ; Buser Say ; Scott Sanner
COMMENTS:       36 pages
HIGHLIGHT:      In this paper, we introduce two types of nonlinear planning methods that can leverage deep neural network learned transition models: Hybrid Deep MILP Planner (HD-MILP-Plan) and Tensorflow Planner (TF-Plan).

65, TITLE:      Deep Entity Matching with Pre-Trained Language Models
http://arxiv.org/abs/2004.00584
AUTHORS:        Yuliang Li ; Jinfeng Li ; Yoshihiko Suhara ; AnHai Doan ; Wang-Chiew Tan
COMMENTS:       In Submission to VLDB 2021
HIGHLIGHT:      We present Ditto, a novel entity matching system based on pre-trained Transformer-based language models.

66, TITLE:      High-Resolution Image Inpainting with Iterative Confidence Feedback and Guided Upsampling
http://arxiv.org/abs/2005.11742
AUTHORS:        Yu Zeng ; Zhe Lin ; Jimei Yang ; Jianming Zhang ; Eli Shechtman ; Huchuan Lu
HIGHLIGHT:      To address this challenge, we propose an iterative inpainting method with a feedback mechanism. Furthermore, to mimic real object removal scenarios, we collect a large object mask dataset and synthesize more realistic training data that better simulates user inputs.

67, TITLE:      SIGMORPHON 2020 Shared Task 0: Typologically Diverse Morphological Inflection
http://arxiv.org/abs/2006.11572
AUTHORS:        Ekaterina Vylomova ; Jennifer White ; Elizabeth Salesky ; Sabrina J. Mielke ; Shijie Wu ; Edoardo Ponti ; Rowan Hall Maudslay ; Ran Zmigrod ; Josef Valvoda ; Svetlana Toldova ; Francis Tyers ; Elena Klyachko ; Ilya Yegorov ; Natalia Krizhanovsky ; Paula Czarnowska ; Irene Nikkarinen ; Andrew Krizhanovsky ; Tiago Pimentel ; Lucas Torroba Hennigen ; Christo Kirov ; Garrett Nicolai ; Adina Williams ; Antonios Anastasopoulos ; Hilaria Cruz ; Eleanor Chodroff ; Ryan Cotterell ; Miikka Silfverberg ; Mans Hulden
COMMENTS:       39 pages, SIGMORPHON
HIGHLIGHT:      SIGMORPHON 2020 Shared Task 0: Typologically Diverse Morphological Inflection

68, TITLE:      Generalized Shortest Path-based Superpixels for Accurate Segmentation of Spherical Images
http://arxiv.org/abs/2004.07394
AUTHORS:        Rémi Giraud ; Rodrigo Borba Pinheiro ; Yannick Berthoumieu
HIGHLIGHT:      In this paper, we introduce a new superpixel method for spherical images called SphSPS (for Spherical Shortest Path-based Superpixels).

69, TITLE:      What Do Compressed Deep Neural Networks Forget?
http://arxiv.org/abs/1911.05248
AUTHORS:        Sara Hooker ; Aaron Courville ; Gregory Clark ; Yann Dauphin ; Andrea Frome
HIGHLIGHT:      Deep neural network pruning and quantization techniques have demonstrated it is possible to achieve high levels of compression with surprisingly little degradation to test set accuracy.

70, TITLE:      Generating Videos of Zero-Shot Compositions of Actions and Objects
http://arxiv.org/abs/1912.02401
AUTHORS:        Megha Nawhal ; Mengyao Zhai ; Andreas Lehrmann ; Leonid Sigal ; Greg Mori
COMMENTS:       Accepted at ECCV'20; Project Page: https://www.sfu.ca/~mnawhal/projects/zs_hoi_generation.html
HIGHLIGHT:      In this paper we develop methods for generating such videos -- making progress toward addressing the important, open problem of video generation in complex scenes.