2020.07.09.txt

==========New Papers==========
1, TITLE:       Artificial Life in Game Mods for Intuitive Evolution Education
http://arxiv.org/abs/2007.03787
AUTHORS:        Anya E. Vostinar ; Barbara Z. Johnson ; Kevin Connors
COMMENTS:       6 pages, 2 figures
HIGHLIGHT:      We propose the use of modifications to commercial games using artificial life techniques to 'stealth teach' about evolution via natural selection, provide a proof-of-concept mod of the game Stardew Valley, and report on its initial reception.

2, TITLE:       Synthetic-to-Real Domain Adaptation for Lane Detection
http://arxiv.org/abs/2007.04023
AUTHORS:        Noa Garnett ; Roy Uziel ; Netalee Efrat ; Dan Levi
HIGHLIGHT:      In this work, we explore learning from abundant, randomly generated synthetic data, together with unlabeled or partially labeled target domain data, instead.

3, TITLE:       A Multi-Level Approach to Waste Object Segmentation
http://arxiv.org/abs/2007.04259
AUTHORS:        Tao Wang ; Yuanzheng Cai ; Lingyu Liang ; Dongyi Ye
COMMENTS:       Paper appears in Sensors 2020, 20(14), 3816
HIGHLIGHT:      We address the problem of localizing waste objects from a color image and an optional depth image, which is a key perception component for robotic interaction with such objects. In addition, we create a new RGBD waste object segmentation dataset, MJU-Waste, that is made public to facilitate future research in this area.

4, TITLE:       Quantifying and Leveraging Predictive Uncertainty for Medical Image Assessment
http://arxiv.org/abs/2007.04258
AUTHORS:        Florin C. Ghesu ; Bogdan Georgescu ; Awais Mansoor ; Youngjin Yoo ; Eli Gibson ; R. S. Vishwanath ; Abishek Balachandran ; James M. Balter ; Yue Cao ; Ramandeep Singh ; Subba R. Digumarthy ; Mannudeep K. Kalra ; Sasa Grbic ; Dorin Comaniciu
COMMENTS:       Under review at Medical Image Analysis
HIGHLIGHT:      To account for this, we propose a system that learns not only the probabilistic estimate for classification, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output.

5, TITLE:       Superpixel Segmentation using Dynamic and Iterative Spanning Forest
http://arxiv.org/abs/2007.04257
AUTHORS:        F. C. Belem ; S. J. F. Guimaraes ; A. X. Falcao
HIGHLIGHT:      In this work, we present Dynamic ISF (DISF) -- a method based on the following steps.

6, TITLE:       Improving Conversational Recommender Systems via Knowledge Graph based Semantic Fusion
http://arxiv.org/abs/2007.04032
AUTHORS:        Kun Zhou ; Wayne Xin Zhao ; Shuqing Bian ; Yuanhang Zhou ; Ji-Rong Wen ; Jingsong Yu
HIGHLIGHT:      To address these issues, we incorporate both word-oriented and entity-oriented knowledge graphs (KG) to enhance the data representations in CRSs, and adopt Mutual Information Maximization to align the word-level and entity-level semantic spaces.

7, TITLE:       Consistency Regularization with Generative Adversarial Networks for Semi-Supervised Image Classification
http://arxiv.org/abs/2007.03844
AUTHORS:        Zexi Chen ; Bharathkumar Ramachandra ; Ranga Raju Vatsavai
COMMENTS:       10 pages, 5 figures
HIGHLIGHT:      In this work, we incorporate the consistency regularization in the vanilla semi-GAN to address this critical limitation.

8, TITLE:       Research on multi-dimensional end-to-end phrase recognition algorithm based on background knowledge
http://arxiv.org/abs/2007.03860
AUTHORS:        Zheng Li ; Gang Tu ; Guang Liu ; Zhi-Qiang Zhan ; Yi-Jian Liu
COMMENTS:       in Chinese language
HIGHLIGHT:      The algorithm can not only introduce background knowledge, recognize all kinds of nested phrases in sentences, but also recognize the dependency between phrases.

9, TITLE:       PaMIR: Parametric Model-Conditioned Implicit Representation for Image-based Human Reconstruction
http://arxiv.org/abs/2007.03858
AUTHORS:        Zerong Zheng ; Tao Yu ; Yebin Liu ; Qionghai Dai
HIGHLIGHT:      To overcome the limitations of regular 3D representations, we propose Parametric Model-Conditioned Implicit Representation (PaMIR), which combines the parametric body model with the free-form deep implicit function.

10, TITLE:      SiENet: Siamese Expansion Network for Image Extrapolation
http://arxiv.org/abs/2007.03851
AUTHORS:        Xiaofeng Zhang ; Feng Chen ; Cailing Wang ; Songsong Wu ; Ming Tao ; Guoping Jiang
HIGHLIGHT:      In this paper, a novel two-stage siamese adversarial model for image extrapolation, named Siamese Expansion Network (SiENet) is proposed.

11, TITLE:      SegFix: Model-Agnostic Boundary Refinement for Segmentation
http://arxiv.org/abs/2007.04269
AUTHORS:        Yuhui Yuan ; Jingyi Xie ; Xilin Chen ; Jingdong Wang
COMMENTS:       ECCV 2020. Project Page: https://github.com/openseg-group/openseg.pytorch
HIGHLIGHT:      We present a model-agnostic post-processing scheme to improve the boundary quality for the segmentation result that is generated by any existing segmentation model.

12, TITLE:      An Improved Upper Bound for SAT
http://arxiv.org/abs/2007.03829
AUTHORS:        Huairui Chu ; Mingyu Xiao ; Zhe Zhang
HIGHLIGHT:      We show that the CNF satisfiability problem can be solved $O^*(1.2226^m)$ time, where $m$ is the number of clauses in the formula, improving the known upper bounds $O^*(1.234^m)$ given by Yamamoto 15 years ago and $O^*(1.239^m)$ given by Hirsch 22 years ago.

13, TITLE:      Dynamic Group Convolution for Accelerating Convolutional Neural Networks
http://arxiv.org/abs/2007.04242
AUTHORS:        Zhuo Su ; Linpu Fang ; Wenxiong Kang ; Dewen Hu ; Matti Pietikäinen ; Li Liu
COMMENTS:       21 pages, 10 figures
HIGHLIGHT:      In this paper, we propose dynamic group convolution (DGC) that adaptively selects which part of input channels to be connected within each group for individual samples on the fly.

14, TITLE:      Predicting the Accuracy of a Few-Shot Classifier
http://arxiv.org/abs/2007.04238
AUTHORS:        Myriam Bontonou ; Louis Béthune ; Vincent Gripon
HIGHLIGHT:      In this paper, we are interested in finding alternatives to answer the question: is my classifier generalizing well to previously unseen data?

15, TITLE:      Delving into the Adversarial Robustness on Face Recognition
http://arxiv.org/abs/2007.04118
AUTHORS:        Xiao Yang ; Dingcheng Yang ; Yinpeng Dong ; Wenjian Yu ; Hang Su ; Jun Zhu
HIGHLIGHT:      Based on our evaluations, we draw several important findings, which are crucial for understanding the adversarial robustness and providing insights for future research on face recognition.

16, TITLE:      Deformable spatial propagation network for depth completion
http://arxiv.org/abs/2007.04251
AUTHORS:        Zheyuan Xu ; Yingfu Wang ; Jian Yao
COMMENTS:       5 pages, 3 figures
HIGHLIGHT:      To tackle this issue, in this paper, we propose a deformable spatial propagation network (DSPN) to adaptively generates different receptive field and affinity matrix for each pixel.

17, TITLE:      A Benchmark of Medical Out of Distribution Detection
http://arxiv.org/abs/2007.04250
AUTHORS:        Tianshi Cao ; Chinwei Huang ; David Yu-Tung Hui ; Joseph Paul Cohen
COMMENTS:       Oral presentation, Uncertainty & Robustness in Deep Learning workshop at ICML. 4 pages, 9 pages total
HIGHLIGHT:      A Benchmark of Medical Out of Distribution Detection

18, TITLE:      Learning Speech Representations from Raw Audio by Joint Audiovisual Self-Supervision
http://arxiv.org/abs/2007.04134
AUTHORS:        Abhinav Shukla ; Stavros Petridis ; Maja Pantic
COMMENTS:       Accepted at the Workshop on Self-supervision in Audio and Speech at ICML 2020
HIGHLIGHT:      We propose a method to learn self-supervised speech representations from the raw audio waveform.

19, TITLE:      Making Adversarial Examples More Transferable and Indistinguishable
http://arxiv.org/abs/2007.03838
AUTHORS:        Junhua Zou ; Zhisong Pan ; Junyang Qiu ; Yexin Duan ; Xin Liu ; Yu Pan
HIGHLIGHT:      To address this problem, we propose an ADAM iterative fast gradient tanh method (AI-FGTM) to generate indistinguishable adversarial examples with high transferability.

20, TITLE:      Language Modeling with Reduced Densities
http://arxiv.org/abs/2007.03834
AUTHORS:        Tai-Danae Bradley ; Yiannis Vlassopoulos
COMMENTS:       19 pages
HIGHLIGHT:      We present a framework for modeling words, phrases, and longer expressions in a natural language using reduced density operators.

21, TITLE:      ISA: An Intelligent Shopping Assistant
http://arxiv.org/abs/2007.03805
AUTHORS:        Tuan Manh Lai ; Trung Bui ; Nedim Lipka
COMMENTS:       6 pages, 5 figures
HIGHLIGHT:      In this paper, we present ISA, a mobile-based intelligent shopping assistant that is designed to improve shopping experience in physical stores.

22, TITLE:      Towards a practical measure of interference for reinforcement learning
http://arxiv.org/abs/2007.03807
AUTHORS:        Vincent Liu ; Adam White ; Hengshuai Yao ; Martha White
COMMENTS:       18 pages
HIGHLIGHT:      In this work, we provide a definition of interference for control in reinforcement learning.

23, TITLE:      A Distilled Model for Tracking and Tracker Fusion
http://arxiv.org/abs/2007.04108
AUTHORS:        Matteo Dunnhofer ; Niki Martinel ; Christian Micheloni
HIGHLIGHT:      In this paper, we unify such goals by proposing a novel tracking methodology that takes advantage of other visual trackers, offline and online.

24, TITLE:      Self-supervised Skull Reconstruction in Brain CT Images with Decompressive Craniectomy
http://arxiv.org/abs/2007.03817
AUTHORS:        Franco Matzkin ; Virginia Newcombe ; Susan Stevenson ; Aneesh Khetani ; Tom Newman ; Richard Digby ; Andrew Stevens ; Ben Glocker ; Enzo Ferrante
COMMENTS:       Accepted for publication in MICCAI 2020
HIGHLIGHT:      We propose and compare alternative self-supervised methods where an encoder-decoder convolutional neural network (CNN) estimates the missing bone flap on post-operative CTs.

25, TITLE:      IOHanalyzer: Performance Analysis for Iterative Optimization Heuristic
http://arxiv.org/abs/2007.03953
AUTHORS:        Hao Wang ; Diederick Vermetten ; Furong Ye ; Carola Doerr ; Thomas Bäck
HIGHLIGHT:      We propose IOHanalyzer, a new software for analyzing the empirical performance of iterative optimization heuristics (IOHs) such as local search algorithms, genetic and evolutionary algorithms, Bayesian optimization algorithms, and similar optimizers.

26, TITLE:      Real-time Semantic Segmentation with Fast Attention
http://arxiv.org/abs/2007.03815
AUTHORS:        Ping Hu ; Federico Perazzi ; Fabian Caba Heilbron ; Oliver Wang ; Zhe Lin ; Kate Saenko ; Stan Sclaroff
COMMENTS:       project page: https://cs-people.bu.edu/pinghu/FANet.html
HIGHLIGHT:      In this paper, we propose a novel architecture that addresses both challenges and achieves state-of-the-art performance for semantic segmentation of high-resolution images and videos in real-time.

27, TITLE:      Expressive Interviewing: A Conversational System for Coping with COVID-19
http://arxiv.org/abs/2007.03819
AUTHORS:        Charles Welch ; Allison Lahnala ; Verónica Pérez-Rosas ; Siqi Shen ; Sarah Seraj ; Larry An ; Kenneth Resnicow ; James Pennebaker ; Rada Mihalcea
HIGHLIGHT:      We introduce \textit{Expressive Interviewing}--an interview-style conversational system that draws on ideas from motivational interviewing and expressive writing.

28, TITLE:      Non-parametric Models for Non-negative Functions
http://arxiv.org/abs/2007.03926
AUTHORS:        Ulysse Marteau-Ferey ; Francis Bach ; Alessandro Rudi
HIGHLIGHT:      In this paper we provide the first model for non-negative functions which benefits from the same good properties of linear models.

29, TITLE:      Winning with Simple Learning Models: Detecting Earthquakes in Groningen, the Netherlands
http://arxiv.org/abs/2007.03924
AUTHORS:        Umair bin Waheed ; Ahmed Shaheen ; Mike Fehler ; Ben Fulcher
HIGHLIGHT:      Here, we revisit the problem of seismic event detection but using a logistic regression model with feature extraction.

30, TITLE:      Operation-Aware Soft Channel Pruning using Differentiable Masks
http://arxiv.org/abs/2007.03938
AUTHORS:        Minsoo Kang ; Bohyung Han
COMMENTS:       ICML 2020
HIGHLIGHT:      We propose a simple but effective data-driven channel pruning algorithm, which compresses deep neural networks in a differentiable way by exploiting the characteristics of operations.

31, TITLE:      BISM: Bytecode-Level Instrumentation for Software Monitoring
http://arxiv.org/abs/2007.03936
AUTHORS:        Chukri Soueidi ; Ali Kassem ; Yliès Falcone
HIGHLIGHT:      The language follows the aspect-oriented programming paradigm by adopting the joinpoint model, advice inlining, and separate instrumentation mechanisms.

32, TITLE:      Self-Supervised Policy Adaptation during Deployment
http://arxiv.org/abs/2007.04309
AUTHORS:        Nicklas Hansen ; Yu Sun ; Pieter Abbeel ; Alexei A. Efros ; Lerrel Pinto ; Xiaolong Wang
COMMENTS:       Project page: https://nicklashansen.github.io/PAD/ , Code: https://github.com/nicklashansen/policy-adaptation-during-deployment
HIGHLIGHT:      Our work explores the use of self-supervision to allow the policy to continue training after deployment without using any rewards.

33, TITLE:      AUSN: Approximately Uniform Quantization by Adaptively Superimposing Non-uniform Distribution for Deep Neural Networks
http://arxiv.org/abs/2007.03903
AUTHORS:        Liu Fangxin ; Zhao Wenbo ; Wang Yanzhi ; Dai Changzhi ; Jiang Li
COMMENTS:       16 pages,6 figures
HIGHLIGHT:      Consequently, we propose a novel quantization method to quantize the weight and activation.

34, TITLE:      Best-First Beam Search
http://arxiv.org/abs/2007.03909
AUTHORS:        Clara Meister ; Ryan Cotterell ; Tim Vieira
COMMENTS:       TACL 2020
HIGHLIGHT:      In this work, we show that standard beam search is a computationally inefficient choice for many decoding tasks; specifically, when the scoring function is a monotonic function in sequence length, other search algorithms can be used to reduce the number of calls to the scoring function (e.g., a neural network), which is often the bottleneck computation.

35, TITLE:      Streaming End-to-End Bilingual ASR Systems with Joint Language Identification
http://arxiv.org/abs/2007.03900
AUTHORS:        Surabhi Punjabi ; Harish Arsikere ; Zeynab Raeesy ; Chander Chandak ; Nikhil Bhave ; Ankish Bansal ; Markus Müller ; Sergio Murillo ; Ariya Rastrow ; Sri Garimella ; Roland Maas ; Mat Hans ; Athanasios Mouchtaris ; Siegfried Kunzmann
HIGHLIGHT:      In this paper, we introduce streaming, end-to-end, bilingual systems that perform both ASR and language identification (LID) using the recurrent neural network transducer (RNN-T) architecture.

36, TITLE:      AutoLR: An Evolutionary Approach to Learning Rate Policies
http://arxiv.org/abs/2007.04223
AUTHORS:        Pedro Carvalho ; Nuno Lourenço ; Filipe Assunção ; Penousal Machado
HIGHLIGHT:      This work presents AutoLR, a framework that evolves Learning Rate Schedulers for a specific Neural Network Architecture using Structured Grammatical Evolution.

37, TITLE:      Dung's semantics satisfy attack removal monotonicity
http://arxiv.org/abs/2007.04221
AUTHORS:        Leila Amgoud ; Srdjan Vesic
HIGHLIGHT:      We show that preferred, stable, complete, and grounded semantics satisfy attack removal monotonicity.

38, TITLE:      Labelling imaging datasets on the basis of neuroradiology reports: a validation study
http://arxiv.org/abs/2007.04226
AUTHORS:        David A. Wood ; Sina Kafiabadi ; Aisha Al Busaidi ; Emily Guilhem ; Jeremy Lynch ; Matthew Townend ; Antanas Montvila ; Juveria Siddiqui ; Naveen Gadapa ; Matthew Benger ; Gareth Barker ; Sebastian Ourselin ; James H. Cole ; Thomas C. Booth
HIGHLIGHT:      In this work, we draw on the experience of a team of neuroradiologists who labelled over 5000 MRI neuroradiology reports as part of a project to build a dedicated deep learning-based neuroradiology report classifier.

39, TITLE:      The Scattering Compositional Learner: Discovering Objects, Attributes, Relationships in Analogical Reasoning
http://arxiv.org/abs/2007.04212
AUTHORS:        Yuhuai Wu ; Honghua Dong ; Roger Grosse ; Jimmy Ba
HIGHLIGHT:      In this work, we focus on an analogical reasoning task that contains rich compositional structures, Raven's Progressive Matrices (RPM).

40, TITLE:      Analysis of Predictive Coding Models for Phonemic Representation Learning in Small Datasets
http://arxiv.org/abs/2007.04205
AUTHORS:        María Andrea Cruz Blandón ; Okko Räsänen
COMMENTS:       7 pages, 5 figures, 5 tables. Accepted paper at the workshop on Self-supervision in Audio and Speech at ICML 2020
HIGHLIGHT:      The present study investigates the behaviour of two predictive coding models, Autoregressive Predictive Coding and Contrastive Predictive Coding, in a phoneme discrimination task (ABX task) for two languages with different dataset sizes.

41, TITLE:      A Natural Actor-Critic Algorithm with Downside Risk Constraints
http://arxiv.org/abs/2007.04203
AUTHORS:        Thomas Spooner ; Rahul Savani
COMMENTS:       14 pages, 5 figures
HIGHLIGHT:      In this paper, we study prediction and control with aversion to downside risk which we gauge by the lower partial moment of the return.

42, TITLE:      Auto-MAP: A DQN Framework for Exploring Distributed Execution Plans for DNN Workloads
http://arxiv.org/abs/2007.04069
AUTHORS:        Siyu Wang ; Yi Rong ; Shiqing Fan ; Zhen Zheng ; LanSong Diao ; Guoping Long ; Jun Yang ; Xiaoyong Liu ; Wei Lin
HIGHLIGHT:      In this paper, we propose Auto-MAP, a framework for exploring distributed execution plans for DNN workloads, which can automatically discovering fast parallelization strategies through reinforcement learning on IR level of deep learning models.

43, TITLE:      Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence
http://arxiv.org/abs/2007.04068
AUTHORS:        Shakir Mohamed ; Marie-Therese Png ; William Isaac
COMMENTS:       28 Pages. Accepted, to appear in: Philosophy and Technology (405), Springer. Submitted 16 January, Accepted 26 May 2020
HIGHLIGHT:      This paper explores the important role of critical science, and in particular of post-colonial and decolonial theories, in understanding and shaping the ongoing advances in artificial intelligence.

44, TITLE:      Learning Neural Textual Representations for Citation Recommendation
http://arxiv.org/abs/2007.04070
AUTHORS:        Binh Thanh Kieu ; Inigo Jauregi Unanue ; Son Bao Pham ; Hieu Xuan Phan ; Massimo Piccardi
COMMENTS:       Accepted in ICPR 2020
HIGHLIGHT:      For this reason, in this paper we propose a novel approach to citation recommendation which leverages a deep sequential representation of the documents (Sentence-BERT) cascaded with Siamese and triplet networks in a submodular scoring function.

45, TITLE:      Single-Frame based Deep View Synchronization for Unsynchronized Multi-Camera Surveillance
http://arxiv.org/abs/2007.03891
AUTHORS:        Qi Zhang ; Antoni B. Chan
COMMENTS:       12 pages
HIGHLIGHT:      To handle the issue of unsynchronized multi-cameras, in this paper, we propose a synchronization model that works in conjunction with existing DNN-based multi-view models, thus avoiding the redesign of the whole model.

46, TITLE:      When Perspective Comes for Free: Improving Depth Prediction with Camera Pose Encoding
http://arxiv.org/abs/2007.03887
AUTHORS:        Yunhan Zhao ; Shu Kong ; Charless Fowlkes
HIGHLIGHT:      To address this challenge, we propose a factored approach that estimates pose first, followed by a conditional depth estimation model that takes an encoding of the camera pose prior (CPP) as input.

47, TITLE:      Automatic Detection of Sexist Statements Commonly Used at the Workplace
http://arxiv.org/abs/2007.04181
AUTHORS:        Dylan Grosz ; Patricia Conde-Cespedes
COMMENTS:       Published at the PAKDD 2020 Workshop on Learning Data Representation for Clustering
HIGHLIGHT:      In this paper we present a dataset of sexist statements that are more likely to be said in the workplace as well as a deep learning model that can achieve state-of-the art results.

48, TITLE:      BS4NN: Binarized Spiking Neural Networks with Temporal Coding and Learning
http://arxiv.org/abs/2007.04039
AUTHORS:        Saeed Reza Kheradpisheh ; Maryam Mirsadeghi ; Timothée Masquelier
HIGHLIGHT:      We recently proposed the S4NN algorithm, essentially an adaptation of backpropagation to multilayer spiking neural networks that use simple non-leaky integrate-and-fire neurons and a form of temporal coding known as time-to-first-spike coding.

49, TITLE:      NVAE: A Deep Hierarchical Variational Autoencoder
http://arxiv.org/abs/2007.03898
AUTHORS:        Arash Vahdat ; Jan Kautz
COMMENTS:       Some images are downsized to meet arXiv requirements. Check https://arash-vahdat.github.io/NVAE_arxiv.pdf for a high-resolution version (24 MB)
HIGHLIGHT:      We propose Nouveau VAE (NVAE), a deep hierarchical VAE built for image generation using depth-wise separable convolutions and batch normalization.

50, TITLE:      A study of Neural networks point source extraction on simulated Fermi/LAT Telescope images
http://arxiv.org/abs/2007.04295
AUTHORS:        Mariia Drozdova ; Anton Broilovskiy ; Andrey Ustyuzhanin ; Denys Malyshev
COMMENTS:       Accepted to Astronomische Nachrichten
HIGHLIGHT:      We present a method for point sources extraction using a convolution neural network (CNN) trained on our own artificial data set which imitates images from the Fermi Large Area Telescope.

51, TITLE:      KIT MOMA: A Mobile Machines Dataset
http://arxiv.org/abs/2007.04198
AUTHORS:        Yusheng Xiang ; Hongzhe Wang ; Tianqing Su ; Ruoyu Li ; Christine Brach ; Samuel S. Mao ; Marcus Geimer
COMMENTS:       15 pages; 17 Figures
HIGHLIGHT:      To address the problem, we publish the KIT MOMA dataset, including eight classes of commonly used mobile machines, which can be used as a benchmark to evaluate the SOTA algorithms to detect mobile construction machines.

52, TITLE:      Marginal loss and exclusion loss for partially supervised multi-organ segmentation
http://arxiv.org/abs/2007.03868
AUTHORS:        Gonglei Shi ; Li Xiao ; Yang Chen ; S. Kevin Zhou
HIGHLIGHT:      In this paper, we investigate how to learn a single multi-organ segmentation network from a union of such datasets.

53, TITLE:      On the Complexity of Horn and Krom Fragments of Second-Order Boolean Logic
http://arxiv.org/abs/2007.03867
AUTHORS:        Miika Hannula ; Juha Kontinen ; Martin Lück ; Jonni Virtema
HIGHLIGHT:      We consider two types of restriction of this logic: 1) restrictions to term constructions, 2) restrictions to the form of the Boolean matrix.

54, TITLE:      Low-dimensional Manifold Constrained Disentanglement Network for Metal Artifact Reduction
http://arxiv.org/abs/2007.03882
AUTHORS:        Chuang Niu ; Wenxiang Cong ; Fenglei Fan ; Hongming Shan ; Mengzhou Li ; Jimin Liang ; Ge Wang
HIGHLIGHT:      To overcome these problems, here we propose a low-dimensional manifold (LDM) constrained disentanglement network (DN), leveraging the image characteristics that the patch manifold is generally low-dimensional.

55, TITLE:      KQA Pro: A Large Diagnostic Dataset for Complex Question Answering over Knowledge Base
http://arxiv.org/abs/2007.03875
AUTHORS:        Jiaxin Shi ; Shulin Cao ; Liangming Pan ; Yutong Xiang ; Lei Hou ; Juanzi Li ; Hanwang Zhang ; Bin He
HIGHLIGHT:      To this end, we introduce KQA Pro, a large-scale dataset for Complex KBQA.

56, TITLE:      Robust Re-Identification by Multiple Views Knowledge Distillation
http://arxiv.org/abs/2007.04174
AUTHORS:        Angelo Porrello ; Luca Bergamini ; Simone Calderara
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      In this work, we devise a training strategy that allows the transfer of a superior knowledge, arising from a set of views depicting the target object.

57, TITLE:      Fine-grained Vibration Based Sensing Using a Smartphone
http://arxiv.org/abs/2007.03874
AUTHORS:        Kamran Ali ; Alex X. Liu
HIGHLIGHT:      In this paper, we propose VibroTag, a robust and practical vibration based sensing scheme that works with smartphones with different hardware, can extract fine-grained vibration signatures of different surfaces, and is robust to environmental noise and hardware based irregularities. We implemented VibroTag on two different Android phones and evaluated in multiple different environments where we collected data from 4 individuals for 5 to 20 consecutive days.

58, TITLE:      Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
http://arxiv.org/abs/2007.04178
AUTHORS:        Junsuk Choe ; Seong Joon Oh ; Sanghyuk Chun ; Zeynep Akata ; Hyunjung Shim
COMMENTS:       TPAMI submission. First two authors contributed equally. This is a journal extension of our CVPR 2020 paper arXiv:2001.07437. Code: https://github.com/clovaai/wsolevaluation
HIGHLIGHT:      In this paper, we argue that WSOL task is ill-posed with only image-level labels, and propose a new evaluation protocol where full supervision is limited to only a small held-out set not overlapping with the test set.

59, TITLE:      PathGAN: Local Path Planning with Generative Adversarial Networks
http://arxiv.org/abs/2007.03877
AUTHORS:        Dooseop Choi ; Seung-jun Han ; Kyoungwook Min ; Jeongdan Choi
HIGHLIGHT:      Targeting autonomous driving without High-Definition maps, we present a model capable of generating multiple plausible paths from sensory inputs for autonomous vehicles. Finally, we introduce ETRIDriving, the dataset for autonomous driving where the recorded sensory data is labeled with discrete high-level driving actions, and demonstrate the-state-of-the-art performances of the proposed model on ETRIDriving in terms of the accuracy and diversity.

60, TITLE:      TripMD: Driving patterns investigation via Motif Analysis
http://arxiv.org/abs/2007.03727
AUTHORS:        Maria Inê Silva ; Roberto Henriques
COMMENTS:       10 pages, 6 figures
HIGHLIGHT:      In this paper, we propose TripMD, a system that extracts the most relevant driving patterns from sensor recordings (such as acceleration) and provides a visualization that allows for an easy investigation.

61, TITLE:      Audio-Visual Understanding of Passenger Intents for In-Cabin Conversational Agents
http://arxiv.org/abs/2007.03876
AUTHORS:        Eda Okur ; Shachi H Kumar ; Saurav Sahay ; Lama Nachman
COMMENTS:       ACL 2020 - Second Grand-Challenge and Workshop on Multimodal Language (Challenge-HML)
HIGHLIGHT:      In this work, we discuss the benefits of a multimodal understanding of in-cabin utterances by incorporating verbal/language input together with the non-verbal/acoustic and visual clues from inside and outside the vehicle.

62, TITLE:      Combating Domain Shift with Self-Taught Labeling
http://arxiv.org/abs/2007.04171
AUTHORS:        Jian Liang ; Dapeng Hu ; Jiashi Feng
HIGHLIGHT:      We present a novel method to combat domain shift when adapting classification models trained on one domain to other new domains with few or no target labels.

63, TITLE:      An exploration of the influence of path choice in game-theoretic attribution algorithms
http://arxiv.org/abs/2007.04169
AUTHORS:        Geoff Ward ; Sean Kamkar ; Jay Budzik
COMMENTS:       21 pages, 23 figures, submitted to JMLR 7/7/2020
HIGHLIGHT:      We compare machine learning explainability methods based on the theory of atomic (Shapley, 1953) and infinitesimal (Aumann and Shapley, 1974) games, in a theoretical and experimental investigation into how the model and choice of integration path can influence the resulting feature attributions.

64, TITLE:      Spatio-Temporal Scene Graphs for Video Dialog
http://arxiv.org/abs/2007.03848
AUTHORS:        Shijie Geng ; Peng Gao ; Chiori Hori ; Jonathan Le Roux ; Anoop Cherian
HIGHLIGHT:      To this end, we propose a novel spatio-temporal scene graph representation (STSGR) modeling fine-grained information flows within videos.

65, TITLE:      Responsive Safety in Reinforcement Learning by PID Lagrangian Methods
http://arxiv.org/abs/2007.03964
AUTHORS:        Adam Stooke ; Joshua Achiam ; Pieter Abbeel
COMMENTS:       ICML 2020
HIGHLIGHT:      We address this shortcoming by proposing a novel Lagrange multiplier update method that utilizes derivatives of the constraint function.

66, TITLE:      Cross-lingual Inductive Transfer to Detect Offensive Language
http://arxiv.org/abs/2007.03771
AUTHORS:        Kartikey Pant ; Tanvi Dadu
COMMENTS:       Accepted at OffenseEval 2020 to be held at COLING 2020
HIGHLIGHT:      In this work, we introduce a cross-lingual inductive approach to identify the offensive language in tweets using the contextual word embedding \textit{XLM-RoBERTa} (XLM-R).

67, TITLE:      SLAP: Improving Physical Adversarial Examples with Short-Lived Adversarial Perturbations
http://arxiv.org/abs/2007.04137
AUTHORS:        Giulio Lovisotto ; Henry Turner ; Ivo Sluganovic ; Martin Strohmeier ; Ivan Martinovic
COMMENTS:       13 pages
HIGHLIGHT:      In this paper, we propose Short-Lived Adversarial Perturbations (SLAP), a novel technique that allows adversaries to realize robust, dynamic real-world AE from a distance.

68, TITLE:      Adaptive 3D Face Reconstruction from a Single Image
http://arxiv.org/abs/2007.03979
AUTHORS:        Kun Li ; Jing Yang ; Nianhong Jiao ; Jinsong Zhang ; Yu-Kun Lai
HIGHLIGHT:      In this paper, we propose a novel joint 2D and 3D optimization method to adaptively reconstruct 3D face shapes from a single image, which combines the depths of 3D landmarks to solve the uncertain detections of invisible landmarks.

69, TITLE:      Evaluating German Transformer Language Models with Syntactic Agreement Tests
http://arxiv.org/abs/2007.03765
AUTHORS:        Karolina Zaczynska ; Nils Feldhus ; Robert Schwarzenberg ; Aleksandra Gabryszak ; Sebastian Möller
COMMENTS:       SwissText + KONVENS 2020
HIGHLIGHT:      In this work, we analyse German TLMs.

70, TITLE:      Detection as Regression: Certified Object Detection by Median Smoothing
http://arxiv.org/abs/2007.03730
AUTHORS:        Ping-yeh Chiang ; Michael J. Curry ; Ahmed Abdelkader ; Aounon Kumar ; John Dickerson ; Tom Goldstein
HIGHLIGHT:      We start by presenting a reduction from object detection to a regression problem.

71, TITLE:      A Quantum Finite Automata Approach to Modeling the Chemical Reactions
http://arxiv.org/abs/2007.03976
AUTHORS:        Amandeep Singh Bhatia ; Shenggen Zheng
COMMENTS:       8 figures, 13 pages
HIGHLIGHT:      In this paper, we have modeled chemical reactions using two-way quantum finite automata, which are halted in linear time.

72, TITLE:      A Free Viewpoint Portrait Generator with Dynamic Styling
http://arxiv.org/abs/2007.03780
AUTHORS:        Anpei Chen ; Ruiyang Liu ; Ling Xie ; Jingyi Yu
HIGHLIGHT:      Therefore, we propose to decompose the generation space into two subspaces: geometric and texture space.

73, TITLE:      README: REpresentation learning by fairness-Aware Disentangling MEthod
http://arxiv.org/abs/2007.03775
AUTHORS:        Sungho Park ; Dohyung Kim ; Sunhee Hwang ; Hyeran Byun
COMMENTS:       8 pages, 3 figures
HIGHLIGHT:      In this paper, we design Fairness-aware Disentangling Variational AutoEncoder (FD-VAE) for fair representation learning.

74, TITLE:      The curious case of developmental BERTology: On sparsity, transfer learning, generalization and the brain
http://arxiv.org/abs/2007.03774
AUTHORS:        Xin Wang
COMMENTS:       9 pages, 5 figures, 1 table
HIGHLIGHT:      In this essay, we explore a point of intersection between deep learning and neuroscience, through the lens of large language models, transfer learning and network compression.

75, TITLE:      3D Shape Reconstruction from Vision and Touch
http://arxiv.org/abs/2007.03778
AUTHORS:        Edward J. Smith ; Roberto Calandra ; Adriana Romero ; Georgia Gkioxari ; David Meger ; Jitendra Malik ; Michal Drozdzal
COMMENTS:       Submitted for review
HIGHLIGHT:      In this paper, we study this problem and present an effective chart-based approach to fusing vision and touch, which leverages advances in graph convolutional networks. To do so, we introduce a dataset of simulated touch and vision signals from the interaction between a robotic hand and a large array of 3D objects.

76, TITLE:      Placepedia: Comprehensive Place Understanding with Multi-Faceted Annotations
http://arxiv.org/abs/2007.03777
AUTHORS:        Huaiyi Huang ; Yuqi Zhang ; Qingqiu Huang ; Zhengkui Guo ; Ziwei Liu ; Dahua Lin
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work, we contribute Placepedia, a large-scale place dataset with more than 35M photos from 240K unique places.

77, TITLE:      Deep Reinforcement Learning and its Neuroscientific Implications
http://arxiv.org/abs/2007.03750
AUTHORS:        Matthew Botvinick ; Jane X. Wang ; Will Dabney ; Kevin J. Miller ; Zeb Kurth-Nelson
COMMENTS:       22 pages, 5 figures
HIGHLIGHT:      In the present review, we provide a high-level introduction to deep RL, discuss some of its initial applications to neuroscience, and survey its wider implications for research on brain and behavior, concluding with a list of opportunities for next-stage research.

78, TITLE:      Learning Efficient Search Approximation in Mixed Integer Branch and Bound
http://arxiv.org/abs/2007.03948
AUTHORS:        Kaan Yilmaz ; Neil Yorke-Smith
HIGHLIGHT:      We present an offline method to learn such a policy in two settings: one that is approximate by committing to pruning of nodes; one that is exact and backtracks from a leaf to use a different strategy.

79, TITLE:      Resonator networks for factoring distributed representations of data structures
http://arxiv.org/abs/2007.03748
AUTHORS:        E. Paxon Frady ; Spencer Kent ; Bruno A. Olshausen ; Friedrich T. Sommer
COMMENTS:       20 pages, 5 figures, to appear in Neural Computation 2020 with companion paper: arXiv:1906.11684
HIGHLIGHT:      In particular, we propose an efficient solution to a hard combinatorial search problem that arises when decoding elements of a VSA data structure: the factorization of products of multiple code vectors.

80, TITLE:      Reconciling Causality and Statistics
http://arxiv.org/abs/2007.03940
AUTHORS:        Pirmin Lemberger ; Denis Oblin
COMMENTS:       22 pages, 14 figures
HIGHLIGHT:      The aim of this pedagogical paper is to present their ideas and methods in a compact and self-contained fashion with concrete business examples as illustrations.

81, TITLE:      Remix: Rebalanced Mixup
http://arxiv.org/abs/2007.03943
AUTHORS:        Hsin-Ping Chou ; Shih-Chieh Chang ; Jia-Yu Pan ; Wei Wei ; Da-Cheng Juan
COMMENTS:       14 pages, 4 figures
HIGHLIGHT:      In this work, we propose a new regularization technique, Remix, that relaxes Mixup's formulation and enables the mixing factors of features and labels to be disentangled.

82, TITLE:      Generalizing Tensor Decomposition for N-ary Relational Knowledge Bases
http://arxiv.org/abs/2007.03988
AUTHORS:        Yu Liu ; Quanming Yao ; Yong Li
COMMENTS:       WWW 2020
HIGHLIGHT:      To generalize tensor decomposition for n-ary relational KBs, in this work, we propose GETD, a generalized model based on Tucker decomposition and Tensor Ring decomposition.

83, TITLE:      Near Optimal Provable Uniform Convergence in Off-Policy Evaluation for Reinforcement Learning
http://arxiv.org/abs/2007.03760
AUTHORS:        Ming Yin ; Yu Bai ; Yu-Xiang Wang
COMMENTS:       Appendix included
HIGHLIGHT:      In this paper, we consider this new question and reveal the comprehensive relationship between OPE and offline learning for the first time.

84, TITLE:      MCU-Net: A framework towards uncertainty representations for decision support system patient referrals in healthcare contexts
http://arxiv.org/abs/2007.03995
AUTHORS:        Nabeel Seedat
COMMENTS:       4 pages, 4 figures, Accepted to ICML 2020 Workshop on Uncertainty & Robustness in Deep Learning & Machine Learning for Global Health
HIGHLIGHT:      Thus, we present a framework of uncertainty representation evaluated for medical image segmentation, using MCU-Net which combines a U-Net with Monte Carlo Dropout, evaluated with four different uncertainty metrics.

85, TITLE:      Designing and Training of A Dual CNN for Image Denoising
http://arxiv.org/abs/2007.03951
AUTHORS:        Chunwei Tian ; Yong Xu ; Wangmeng Zuo ; Bo Du ; Chia-Wen Lin ; David Zhang
HIGHLIGHT:      In this paper, we propsoed a Dual denoising Network (DudeNet) to recover a clean image.

86, TITLE:      Guidestar-free image-guided wavefront-shaping
http://arxiv.org/abs/2007.03956
AUTHORS:        Tomer Yeminy ; Ori Katz
HIGHLIGHT:      Here, we present a new concept, image-guided wavefront-shaping, allowing non-invasive, guidestar-free, widefield, incoherent imaging through highly scattering layers, without illumination control.



==========Updates to Previous Papers==========
1, TITLE:       You Autocomplete Me: Poisoning Vulnerabilities in Neural Code Completion
http://arxiv.org/abs/2007.02220
AUTHORS:        Roei Schuster ; Congzheng Song ; Eran Tromer ; Vitaly Shmatikov
HIGHLIGHT:      We demonstrate that neural code autocompleters are vulnerable to data- and model-poisoning attacks.

2, TITLE:       Naive-Student: Leveraging Semi-Supervised Learning in Video Sequences for Urban Scene Segmentation
http://arxiv.org/abs/2005.10266
AUTHORS:        Liang-Chieh Chen ; Raphael Gontijo Lopes ; Bowen Cheng ; Maxwell D. Collins ; Ekin D. Cubuk ; Barret Zoph ; Hartwig Adam ; Jonathon Shlens
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this work, we ask if we may leverage semi-supervised learning in unlabeled video sequences and extra images to improve the performance on urban scene segmentation, simultaneously tackling semantic, instance, and panoptic segmentation.

3, TITLE:       GRNet: Gridding Residual Network for Dense Point Cloud Completion
http://arxiv.org/abs/2006.03761
AUTHORS:        Haozhe Xie ; Hongxun Yao ; Shangchen Zhou ; Jiageng Mao ; Shengping Zhang ; Wenxiu Sun
COMMENTS:       ECCV 2020
HIGHLIGHT:      To solve this problem, we introduce 3D grids as intermediate representations to regularize unordered point clouds.

4, TITLE:       Lost in translation: Exposing hidden compiler optimization opportunities
http://arxiv.org/abs/1903.11397
AUTHORS:        Kyriakos Georgiou ; Zbigniew Chamski ; Andres Amaya Garcia ; David May ; Kerstin Eder
COMMENTS:       31 pages, 7 figures, 2 table. arXiv admin note: text overlap with arXiv:1802.09845
HIGHLIGHT:      As a case study, we demonstrate how the insights from our approach enabled us to identify and remove a significant shortcoming of the CFG simplification pass of the LLVM v6.0.1 compiler.

5, TITLE:       Lightweight Temporal Self-Attention for Classifying Satellite Image Time Series
http://arxiv.org/abs/2007.00586
AUTHORS:        Vivien Sainte Fare Garnot ; Loic Landrieu
HIGHLIGHT:      Building on recent work employing multi-headed self-attention mechanisms to classify remote sensing time sequences, we propose a modification of the Temporal Attention Encoder.

6, TITLE:       Domain Confusion with Self Ensembling for Unsupervised Adaptation
http://arxiv.org/abs/1810.04472
AUTHORS:        Jiawei Wang ; Zhaoshui He ; Chengjian Feng ; Zhouping Zhu ; Qinzhuang Lin ; Jun Lv ; Shengli Xie
COMMENTS:       The expression is ambiguous, which is not convenient for readers to understand, and in today's view, the conclusion of the paper is of little significance, so it is no longer open
HIGHLIGHT:      The experiments verified that our proposed approach can offer better performance than empirical art in a variety of unsupervised domain adaptation benchmarks.

7, TITLE:       Predicting Temporal Sets with Deep Neural Networks
http://arxiv.org/abs/2006.11483
AUTHORS:        Le Yu ; Leilei Sun ; Bowen Du ; Chuanren Liu ; Hui Xiong ; Weifeng Lv
COMMENTS:       9 pages, 6 figures, Proceedings of the 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '2020)
HIGHLIGHT:      In this paper, we propose an integrated solution based on the deep neural networks for temporal sets prediction.

8, TITLE:       Population-based Gradient Descent Weight Learning for Graph Coloring Problems
http://arxiv.org/abs/1909.02261
AUTHORS:        Olivier Goudet ; Béatrice Duval ; Jin-Kao Hao
HIGHLIGHT:      In this work, we present a general population-based weight learning framework for solving graph coloring problems.

9, TITLE:       Working Memory Graphs
http://arxiv.org/abs/1911.07141
AUTHORS:        Ricky Loynd ; Roland Fernandez ; Asli Celikyilmaz ; Adith Swaminathan ; Matthew Hausknecht
COMMENTS:       9 pages, 6 figures, 7 page appendix
HIGHLIGHT:      We present the Working Memory Graph (WMG), an agent that employs multi-head self-attention to reason over a dynamic set of vectors representing observed and recurrent state.

10, TITLE:      Descriptive complexity of real computation and probabilistic independence logic
http://arxiv.org/abs/2003.00644
AUTHORS:        Miika Hannula ; Juha Kontinen ; Jan Van den Bussche ; Jonni Virtema
HIGHLIGHT:      We introduce a novel variant of BSS machines called Separate Branching BSS machines (S-BSS in short) and develop a Fagin-type logical characterisation for languages decidable in non-deterministic polynomial time by S-BSS machines.

11, TITLE:      Delta Schema Network in Model-based Reinforcement Learning
http://arxiv.org/abs/2006.09950
AUTHORS:        Andrey Gorodetskiy ; Alexandra Shlychkova ; Aleksandr I. Panov
COMMENTS:       Published at the AGI 2020 conference
HIGHLIGHT:      In the paper we are expanding the schema networks method which allows to extract the logical relationships between objects and actions from the environment data.

12, TITLE:      Flexible and Efficient Long-Range Planning Through Curious Exploration
http://arxiv.org/abs/2004.10876
AUTHORS:        Aidan Curtis ; Minjian Xin ; Dilip Arumugam ; Kevin Feigelis ; Daniel Yamins
HIGHLIGHT:      Here, we propose the Curious Sample Planner (CSP), which fuses elements of TAMP and DRL by combining a curiosity-guided sampling strategy with imitation learning to accelerate planning.

13, TITLE:      Deep Learning for Anomaly Detection: A Review
http://arxiv.org/abs/2007.02500
AUTHORS:        Guansong Pang ; Chunhua Shen ; Longbing Cao ; Anton van den Hengel
COMMENTS:       Survey paper, 36 pages, 180 references, 2 figures, 3 tables
HIGHLIGHT:      We review their key intuitions, objective functions, underlying assumptions, advantages and disadvantages, and discuss how they address the aforementioned challenges.

14, TITLE:      A Few-Shot Sequential Approach for Object Counting
http://arxiv.org/abs/2007.01899
AUTHORS:        Negin Sokhandan ; Pegah Kamousi ; Alejandro Posada ; Eniola Alese ; Negar Rostamzadeh
HIGHLIGHT:      In this work, we address the problem of few-shot multi-class object counting with point-level annotations. In addition, we introduce a new dataset that is specifically designed for weakly supervised multi-class object counting/detection and contains considerably different classes and distribution of number of classes/instances per image compared to the existing datasets.

15, TITLE:      RGBT Salient Object Detection: A Large-scale Dataset and Benchmark
http://arxiv.org/abs/2007.03262
AUTHORS:        Zhengzheng Tu ; Yan Ma ; Zhun Li ; Chenglong Li ; Jieming Xu ; Yongtao Liu
COMMENTS:       12 pages, 10 figures
HIGHLIGHT:      With this dataset, we propose a powerful baseline approach, which extracts multi-level features within each modality and aggregates these features of all modalities with the attention mechanism, for accurate RGBT salient object detection.

16, TITLE:      Simplicial Complex based Point Correspondence between Images warped onto Manifolds
http://arxiv.org/abs/2007.02381
AUTHORS:        Charu Sharma ; Manohar Kaul
COMMENTS:       Accepted at ECCV 2020
HIGHLIGHT:      In this paper, we pose the assignment problem as finding a bijective map between two graph induced simplicial complexes, which are higher-order analogues of graphs. We propose a constrained quadratic assignment problem (QAP) that matches each p-skeleton of the simplicial complexes, iterating from the highest to the lowest dimension.

17, TITLE:      Robust Market Making via Adversarial Reinforcement Learning
http://arxiv.org/abs/2003.01820
AUTHORS:        Thomas Spooner ; Rahul Savani
COMMENTS:       7 pages, 3 figures; IJCAI-PRICAI '20 Conference Proceedings
HIGHLIGHT:      We show that adversarial reinforcement learning (ARL) can be used to produce market marking agents that are robust to adversarial and adaptively-chosen market conditions.

18, TITLE:      Beetle Swarm Optimization Algorithm:Theory and Application
http://arxiv.org/abs/1808.00206
AUTHORS:        Tiantian Wang ; Long Yang
HIGHLIGHT:      In this paper, a new meta-heuristic algorithm, called beetle swarm optimization algorithm, is proposed by enhancing the performance of swarm optimization through beetle foraging principles.

19, TITLE:      Massively Multilingual ASR: 50 Languages, 1 Model, 1 Billion Parameters
http://arxiv.org/abs/2007.03001
AUTHORS:        Vineel Pratap ; Anuroop Sriram ; Paden Tomasello ; Awni Hannun ; Vitaliy Liptchinsky ; Gabriel Synnaeve ; Ronan Collobert
HIGHLIGHT:      We compare three variants of multilingual training from a single joint model without knowing the input language, to using this information, to multiple heads (one per language cluster).

20, TITLE:      User Intent Inference for Web Search and Conversational Agents
http://arxiv.org/abs/2005.13808
AUTHORS:        Ali Ahmadvand
COMMENTS:       WSDM2020
HIGHLIGHT:      To address the first topic, I proposed novel models to incorporate entity information and conversation-context clues to predict both topic and intent of the user's utterances.

21, TITLE:      Learning Enriched Features for Real Image Restoration and Enhancement
http://arxiv.org/abs/2003.06792
AUTHORS:        Syed Waqas Zamir ; Aditya Arora ; Salman Khan ; Munawar Hayat ; Fahad Shahbaz Khan ; Ming-Hsuan Yang ; Ling Shao
COMMENTS:       Accepted for publication at ECCV 2020
HIGHLIGHT:      In this paper, we present a novel architecture with the collective goals of maintaining spatially-precise high-resolution representations through the entire network and receiving strong contextual information from the low-resolution representations.

22, TITLE:      Signature-Based Abduction for Expressive Description Logics -- Technical Report
http://arxiv.org/abs/2007.00757
AUTHORS:        Patrick Koopmann ; Warren Del-Pinto ; Sophie Tourret ; Renate A. Schmidt
COMMENTS:       13 pages, 1 figure
HIGHLIGHT:      We present the first complete method solving signature-based abduction for observations expressed in the expressive description logic ALC, which can include TBox and ABox axioms, thereby solving the knowledge base abduction problem.

23, TITLE:      Object-Contextual Representations for Semantic Segmentation
http://arxiv.org/abs/1909.11065
AUTHORS:        Yuhui Yuan ; Xilin Chen ; Jingdong Wang
COMMENTS:       ECCV 2020 Spotlight. Project Page: https://github.com/openseg-group/openseg.pytorch; https://github.com/HRNet/HRNet-Semantic-Segmentation/tree/HRNet-OCR
HIGHLIGHT:      In this paper, we address the semantic segmentation problem with a focus on the context aggregation strategy.

24, TITLE:      Derandomization and absolute reconstruction for sums of powers of linear forms
http://arxiv.org/abs/1912.02021
AUTHORS:        Pascal Koiran ; Mateusz Skomra
COMMENTS:       This version includes a short discussion of Jennrich's algorithm
HIGHLIGHT:      As one of our main results we give an algorithm for the following problem: given a homogeneous polynomial of degree 3, decide whether it can be written as a sum of cubes of linearly independent linear forms with complex coefficients.

25, TITLE:      Adaptive Transformers for Learning Multimodal Representations
http://arxiv.org/abs/2005.07486
AUTHORS:        Prajjwal Bhargava
COMMENTS:       Accepted at ACL SRW 2020. Code can be found here https://github.com/prajjwal1/adaptive_transformer
HIGHLIGHT:      In this work, we extend adaptive approaches to learn more about model interpretability and computational efficiency.

26, TITLE:      A Convolutional Approach to Vertebrae Detection and Labelling in Whole Spine MRI
http://arxiv.org/abs/2007.02606
AUTHORS:        Rhydian Windsor ; Amir Jamaludin ; Timor Kadir ; Rhydian Windsor
COMMENTS:       Accepted full paper to Medical Image Computing and Computer Assisted Intervention 2020. 11 pages plus appendix
HIGHLIGHT:      We propose a novel convolutional method for the detection and identification of vertebrae in whole spine MRIs.

27, TITLE:      Single Path One-Shot Neural Architecture Search with Uniform Sampling
http://arxiv.org/abs/1904.00420
AUTHORS:        Zichao Guo ; Xiangyu Zhang ; Haoyuan Mu ; Wen Heng ; Zechun Liu ; Yichen Wei ; Jian Sun
COMMENTS:       ECCV 2020
HIGHLIGHT:      This work propose a Single Path One-Shot model to address the challenge in the training.

28, TITLE:      Dense Recurrent Neural Networks for Accelerated MRI: History-Cognizant Unrolling of Optimization Algorithms
http://arxiv.org/abs/1912.07197
AUTHORS:        Seyed Amir Hossein Hosseini ; Burhaneddin Yaman ; Steen Moeller ; Mingyi Hong ; Mehmet Akçakaya
HIGHLIGHT:      Recently physics-driven deep learning (DL) methods have been proposed to use neural networks for data-driven regularization.

29, TITLE:      AutoML: A Survey of the State-of-the-Art
http://arxiv.org/abs/1908.00709
AUTHORS:        Xin He ; Kaiyong Zhao ; Xiaowen Chu
COMMENTS:       automated machine learning (AutoML), Submitted to Knowledge Based Systems for review
HIGHLIGHT:      In this paper, we provide a comprehensive and up-to-date review of the state-of-the-art (SOTA) in AutoML.

30, TITLE:      Representation Learning with Fine-grained Patterns
http://arxiv.org/abs/2005.09681
AUTHORS:        Yuanhong Xu ; Qi Qian ; Hao Li ; Rong Jin ; Juhua Hu
HIGHLIGHT:      To mitigate the challenge, we propose an algorithm to learn the fine-grained patterns sufficiently when only super-class labels are available.

31, TITLE:      AlignShift: Bridging the Gap of Imaging Thickness in 3D Anisotropic Volumes
http://arxiv.org/abs/2005.01969
AUTHORS:        Jiancheng Yang ; Yi He ; Xiaoyang Huang ; Jingwei Xu ; Xiaodan Ye ; Guangyu Tao ; Bingbing Ni
COMMENTS:       MICCAI 2020 (early accepted). Camera ready version. Code is available at https://github.com/M3DV/AlignShift
HIGHLIGHT:      We aim at a unified approach for both thin- and thick-slice medical volumes.

32, TITLE:      Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation
http://arxiv.org/abs/2007.01947
AUTHORS:        Guolei Sun ; Wenguan Wang ; Jifeng Dai ; Luc Van Gool
COMMENTS:       Full version of ECCV2020 Oral, CVPR2020 LID workshop Best Paper and LID challenge Track1 winner; website: https://github.com/GuoleiSun/MCIS_wsss
HIGHLIGHT:      This paper studies the problem of learning semantic segmentation from image-level supervision only.

33, TITLE:      Modeling Lost Information in Lossy Image Compression
http://arxiv.org/abs/2006.11999
AUTHORS:        Yaolong Wang ; Mingqing Xiao ; Chang Liu ; Shuxin Zheng ; Tie-Yan Liu
HIGHLIGHT:      In this work, we propose a novel invertible framework called Invertible Lossy Compression (ILC) to largely mitigate the information loss problem.

34, TITLE:      CycleGAN with a Blur Kernel for Deconvolution Microscopy: Optimal Transport Geometry
http://arxiv.org/abs/1908.09414
AUTHORS:        Sungjun Lim ; Hyoungjun Park ; Sang-Eun Lee ; Sunghoe Chang ; Jong Chul Ye
COMMENTS:       This paper is accepted for IEEE Trans. Computational Imaging
HIGHLIGHT:      In this paper, we present a novel unsupervised cycle-consistent generative adversarial network (cycleGAN) with a linear blur kernel, which can be used for both blind- and non-blind image deconvolution.

35, TITLE:      Multimodal Shape Completion via Conditional Generative Adversarial Networks
http://arxiv.org/abs/2003.07717
AUTHORS:        Rundi Wu ; Xuelin Chen ; Yixin Zhuang ; Baoquan Chen
COMMENTS:       Accepted to ECCV 2020 (spotlight). Project page at https://chriswu1997.github.io/files/multimodal-pc/index.html
HIGHLIGHT:      Several deep learning methods have been proposed for completing partial data from shape acquisition setups, i.e., filling the regions that were missing in the shape. Hence, we pose a multi-modal shape completion problem, in which we seek to complete the partial shape with multiple outputs by learning a one-to-many mapping.

36, TITLE:      Unsupervised Learning of Landmarks based on Inter-Intra Subject Consistencies
http://arxiv.org/abs/2004.07936
AUTHORS:        Weijian Li ; Haofu Liao ; Shun Miao ; Le Lu ; Jiebo Luo
COMMENTS:       Accepted to ICPR-20
HIGHLIGHT:      We present a novel unsupervised learning approach to image landmark discovery by incorporating the inter-subject landmark consistencies on facial images.

37, TITLE:      MedDialog: Two Large-scale Medical Dialogue Datasets
http://arxiv.org/abs/2004.03329
AUTHORS:        Xuehai He ; Shu Chen ; Zeqian Ju ; Xiangyu Dong ; Hongchao Fang ; Sicheng Wang ; Yue Yang ; Jiaqi Zeng ; Ruisi Zhang ; Ruoyu Zhang ; Meng Zhou ; Penghui Zhu ; Pengtao Xie
HIGHLIGHT:      To facilitate the research and development of medical dialogue systems, we build two large-scale medical dialogue datasets: MedDialog-EN and MedDialog-CN.

38, TITLE:      Depth-Aware Arbitrary Style Transfer Using Instance Normalization
http://arxiv.org/abs/1906.01123
AUTHORS:        Victor Kitov ; Konstantin Kozlovtsev ; Margarita Mishustina
COMMENTS:       Replacement of the previous version due to the following improvements: depth estimation methods comparison added, better depth estimation network used, transformation to proximity map added with offset and contrast parameters. Dependency on these parameters shown, comparison of AdaIN and proposed method added, user evaluation study completely remade for improved version of the proposed method
HIGHLIGHT:      We propose an extension to this method, allowing depth map preservation by applying variable stylization strength.

39, TITLE:      Enhancing Underexposed Photos using Perceptually Bidirectional Similarity
http://arxiv.org/abs/1907.10992
AUTHORS:        Qing Zhang ; Yongwei Nie ; Lei Zhu ; Chunxia Xiao ; Wei-Shi Zheng
COMMENTS:       Aceepted to IEEE Transactions on Multimedia (TMM)
HIGHLIGHT:      To obtain high-quality results free of these artifacts, we present a novel underexposed photo enhancement approach that is able to maintain the perceptual consistency.

40, TITLE:      Phase Retrieval Using Conditional Generative Adversarial Networks
http://arxiv.org/abs/1912.04981
AUTHORS:        Tobias Uelwer ; Alexander Oberstraß ; Stefan Harmeling
COMMENTS:       Accepted at the 25th International Conference on Pattern Recognition 2020 (ICPR)
HIGHLIGHT:      In this paper, we propose the application of conditional generative adversarial networks to solve various phase retrieval problems.

41, TITLE:      Building Damage Annotation on Post-Hurricane Satellite Imagery Based on Convolutional Neural Networks
http://arxiv.org/abs/1807.01688
AUTHORS:        Quoc Dung Cao ; Youngjun Choe
HIGHLIGHT:      In this paper, we propose to improve the efficiency of building damage assessment by applying image classification algorithms to post-hurricane satellite imagery.

42, TITLE:      Multi-image Super Resolution of Remotely Sensed Images using Residual Feature Attention Deep Neural Networks
http://arxiv.org/abs/2007.03107
AUTHORS:        Francesco Salvetti ; Vittorio Mazzia ; Aleem Khaliq ; Marcello Chiaberge
HIGHLIGHT:      In this context, the presented research proposes a novel residual attention model (RAMS) that efficiently tackles the multi-image super-resolution task, simultaneously exploiting spatial and temporal correlations to combine multiple images.

43, TITLE:      Am I Building a White Box Agent or Interpreting a Black Box Agent?
http://arxiv.org/abs/2007.01187
AUTHORS:        Tom Bewley
COMMENTS:       6 pages
HIGHLIGHT:      The rule extraction literature contains the notion of a fidelity-accuracy dilemma: when building an interpretable model of a black box function, optimising for fidelity is likely to reduce performance on the underlying task, and vice versa.

44, TITLE:      AVP-SLAM: Semantic Visual Mapping and Localization for Autonomous Vehicles in the Parking Lot
http://arxiv.org/abs/2007.01813
AUTHORS:        Tong Qin ; Tongqing Chen ; Yilun Chen ; Qing Su
COMMENTS:       The IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2020
HIGHLIGHT:      In this paper, we exploit robust semantic features to build the map and localize vehicles in parking lots.

45, TITLE:      DeepACEv2: Automated Chromosome Enumeration in Metaphase Cell Images Using Deep Convolutional Neural Networks
http://arxiv.org/abs/1910.11091
AUTHORS:        Chunlong Luo ; Li Xiao ; Tianqi Yu ; Yufan Luo ; Manqing Wang ; Fuhai Yu ; Yinhao Li ; Chan Tian ; Jie Qiao
COMMENTS:       There is some fatal errors and I need to made the change
HIGHLIGHT:      To automate the enumeration process, we develop a chromosome enumeration framework, DeepACEv2, based on the region based object detection scheme.

46, TITLE:      LandCover.ai: Dataset for Automatic Mapping of Buildings, Woodlands and Water from Aerial Imagery
http://arxiv.org/abs/2005.02264
AUTHORS:        Adrian Boguszewski ; Dominik Batorski ; Natalia Ziemba-Jankowska ; Anna Zambrzycka ; Tomasz Dziedzic
HIGHLIGHT:      Here we introduce LandCover.ai (Land Cover from Aerial Imagery) dataset for semantic segmentation.

47, TITLE:      Probabilistic Guarantees for Safe Deep Reinforcement Learning
http://arxiv.org/abs/2005.07073
AUTHORS:        Edoardo Bacci ; David Parker
HIGHLIGHT:      We propose MOSAIC, an algorithm for measuring the safety of deep reinforcement learning agents in stochastic settings.

48, TITLE:      An Integer Programming Approach to Deep Neural Networks with Binary Activation Functions
http://arxiv.org/abs/2007.03326
AUTHORS:        Bubacarr Bah ; Jannis Kurtz
HIGHLIGHT:      We implemented our methods on random and real datasets and show that the heuristic version of the BDNN outperforms classical deep neural networks on the Breast Cancer Wisconsin dataset while performing worse on random data.

49, TITLE:      Public discourse and sentiment during the COVID-19 pandemic: using Latent Dirichlet Allocation for topic modeling on Twitter
http://arxiv.org/abs/2005.08817
AUTHORS:        Jia Xue ; Junxiang Chen ; Chen Chen ; Chengda Zheng ; Sijia Li ; Tingshao Zhu
HIGHLIGHT:      The study aims to understand Twitter users' discourse and psychological reactions to COVID-19.

50, TITLE:      HKR For Handwritten Kazakh & Russian Database
http://arxiv.org/abs/2007.03579
AUTHORS:        Daniyar Nurseitov ; Kairat Bostanbekov ; Daniyar Kurmankhojayev ; Anel Alimova ; Abdelrahman Abdallah
HIGHLIGHT:      In this paper, we present a new Russian and Kazakh database (with about 95% of Russian and 5% of Kazakh words/sentences respectively) for offline handwriting recognition.

51, TITLE:      ReMOTS: Self-Supervised Refining Multi-Object Tracking and Segmentation
http://arxiv.org/abs/2007.03200
AUTHORS:        Fan Yang ; Xin Chang ; Chenyu Dang ; Ziqiang Zheng ; Sakriani Sakti ; Satoshi Nakamura ; Yang Wu
COMMENTS:       4 pages
HIGHLIGHT:      To tackle this issue, we propose a self-supervised refining MOTS (i.e., ReMOTS) framework.

52, TITLE:      Open Set Domain Adaptation with Multi-Classifier Adversarial Network
http://arxiv.org/abs/2007.00384
AUTHORS:        Tasfia Shermin ; Guojun Lu ; Shyh Wei Teng ; Manzur Murshed ; Ferdous Sohel
HIGHLIGHT:      For addressing this specific open set domain adaptation setting, prior research introduces a domain adversarial model with an empirical fixed threshold which lacks at handling false-negative transfers.

53, TITLE:      Fully Hyperbolic Convolutional Neural Networks
http://arxiv.org/abs/1905.10484
AUTHORS:        Keegan Lensink ; Bas Peters ; Eldad Haber
COMMENTS:       21 pages, 9 figures, Updated work to include additional numerical experiments, a section about VAEs and learnable wavelets
HIGHLIGHT:      Motivated by the propagation of signals over physical networks, that are governed by the hyperbolic Telegraph equation, in this work we introduce a fully conservative hyperbolic network for problems with high dimensional input and output.

54, TITLE:      Learning to Count in the Crowd from Limited Labeled Data
http://arxiv.org/abs/2007.03195
AUTHORS:        Vishwanath A. Sindagi ; Rajeev Yasarla ; Deepak Sam Babu ; R. Venkatesh Babu ; Vishal M. Patel
COMMENTS:       Accepted at ECCV 2020
HIGHLIGHT:      In this work, we focus on reducing the annotation efforts by learning to count in the crowd from limited number of labeled samples while leveraging a large pool of unlabeled data.

55, TITLE:      Gradient Origin Networks
http://arxiv.org/abs/2007.02798
AUTHORS:        Sam Bond-Taylor ; Chris G. Willcocks
COMMENTS:       5 pages, 7 figures, fixed missing negative and brackets
HIGHLIGHT:      This paper proposes a new type of implicit generative model that is able to quickly learn a latent representation without an explicit encoder.

56, TITLE:      Quantum computation with indefinite causal structures
http://arxiv.org/abs/1706.09854
AUTHORS:        Mateus Araújo ; Philippe Allard Guérin ; Ämin Baumeler
COMMENTS:       11 + 5 pages, no figures, 16 circuits. Corrected equations (33)-(36)
HIGHLIGHT:      In this paper we show that process matrices correspond to a linear particular case of P-CTCs, and therefore that its computational power is upperbounded by that of PP.

57, TITLE:      Text-based depression detection on sparse data
http://arxiv.org/abs/1904.05154
AUTHORS:        Heinrich Dinkel ; Mengyue Wu ; Kai Yu
HIGHLIGHT:      This work proposes a text-based multi-task BGRU network with pretrained word embeddings to model patients' responses during clinical interviews.

58, TITLE:      A Simple Language Model for Task-Oriented Dialogue
http://arxiv.org/abs/2005.00796
AUTHORS:        Ehsan Hosseini-Asl ; Bryan McCann ; Chien-Sheng Wu ; Semih Yavuz ; Richard Socher
COMMENTS:       22 Pages, 2 figures, 16 tables
HIGHLIGHT:      SimpleTOD is a simple approach to task-oriented dialogue that uses a single causal language model trained on all sub-tasks recast as a single sequence prediction problem.

59, TITLE:      Parameter Sharing is Surprisingly Useful for Multi-Agent Deep Reinforcement Learning
http://arxiv.org/abs/2005.13625
AUTHORS:        Justin K Terry ; Nathaniel Grammel ; Ananth Hari ; Luis Santos
HIGHLIGHT:      We use the MAILP model to show that increasing training centralization arbitrarily mitigates the slowing of convergence due to nonstationarity.

60, TITLE:      Strong Generalization and Efficiency in Neural Programs
http://arxiv.org/abs/2007.03629
AUTHORS:        Yujia Li ; Felix Gimeno ; Pushmeet Kohli ; Oriol Vinyals
HIGHLIGHT:      We study the problem of learning efficient algorithms that strongly generalize in the framework of neural program induction.

61, TITLE:      Graph2Kernel Grid-LSTM: A Multi-Cued Model for Pedestrian Trajectory Prediction by Learning Adaptive Neighborhoods
http://arxiv.org/abs/2007.01915
AUTHORS:        Sirin Haddad ; Siew Kei Lam
HIGHLIGHT:      We present a new perspective to interaction modeling by proposing that pedestrian neighborhoods can become adaptive in design.

62, TITLE:      CNN-based fast source device identification
http://arxiv.org/abs/2001.11847
AUTHORS:        Sara Mandelli ; Davide Cozzolino ; Paolo Bestagini ; Luisa Verdoliva ; Stefano Tubaro
HIGHLIGHT:      In this paper we address the problem of device identification based on sensor noise and propose a fast and accurate solution using convolutional neural networks (CNNs).

63, TITLE:      Learning to Segment Anatomical Structures Accurately from One Exemplar
http://arxiv.org/abs/2007.03052
AUTHORS:        Yuhang Lu ; Weijian Li ; Kang Zheng ; Yirui Wang ; Adam P. Harrison ; Chihung Lin ; Song Wang ; Jing Xiao ; Le Lu ; Chang-Fu Kuo ; Shun Miao
COMMENTS:       MICCAI2020 (Early accept)
HIGHLIGHT:      In this work, we propose a novel contribution of Contour Transformer Network (CTN), a one-shot anatomy segmentor including a naturally built-in human-in-the-loop mechanism.

64, TITLE:      Biomechanics-informed Neural Networks for Myocardial Motion Tracking in MRI
http://arxiv.org/abs/2006.04725
AUTHORS:        Chen Qin ; Shuo Wang ; Chen Chen ; Huaqi Qiu ; Wenjia Bai ; Daniel Rueckert
COMMENTS:       The paper is early accepted by MICCAI 2020
HIGHLIGHT:      In contrast to most of the current approaches which impose explicit regularisation terms such as smoothness, in this paper we propose a novel method that can implicitly learn biomechanics-informed regularisation.

65, TITLE:      RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification
http://arxiv.org/abs/2006.12634
AUTHORS:        Jingtian Peng ; Chang Xiao ; Xun Wei ; Yifan Li
HIGHLIGHT:      We introduce RP2K, a new large-scale retail product dataset for fine-grained image classification.

66, TITLE:      Guided Fine-Tuning for Large-Scale Material Transfer
http://arxiv.org/abs/2007.03059
AUTHORS:        Valentin Deschaintre ; George Drettakis ; Adrien Bousseau
COMMENTS:       Published in Computer Graphics Forum, 39(4); Proceedings of the Eurographics Symposium on Rendering 2020
HIGHLIGHT:      We present a method to transfer the appearance of one or a few exemplar SVBRDFs to a target image representing similar materials.

67, TITLE:      Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control
http://arxiv.org/abs/1910.07972
AUTHORS:        Lukas Hermann ; Max Argus ; Andreas Eitel ; Artemij Amiranashvili ; Wolfram Burgard ; Thomas Brox
COMMENTS:       Accepted at the 2020 IEEE International Conference on Robotics and Automation (ICRA). Project page see https://lmb.informatik.uni-freiburg.de/projects/curriculum/
HIGHLIGHT:      We propose Adaptive Curriculum Generation from Demonstrations (ACGD) for reinforcement learning in the presence of sparse rewards.

68, TITLE:      Undermining User Privacy on Mobile Devices Using AI
http://arxiv.org/abs/1811.11218
AUTHORS:        Berk Gulmezoglu ; Andreas Zankl ; M. Caner Tol ; Saad Islam ; Thomas Eisenbarth ; Berk Sunar
HIGHLIGHT:      In this work, we show that these inference attacks are considerably more practical when combined with advanced AI techniques.

69, TITLE:      On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach
http://arxiv.org/abs/1803.09719
AUTHORS:        Nikolai Smolyanskiy ; Alexey Kamenev ; Stan Birchfield
COMMENTS:       CVPR 2018 Workshop on Autonomous Driving. For video, see https://youtu.be/0FPQdVOYoAU
HIGHLIGHT:      We propose a novel semi-supervised learning approach to training a deep stereo neural network, along with a novel architecture containing a machine-learned argmax layer and a custom runtime (that will be shared publicly) that enables a smaller version of our stereo DNN to run on an embedded GPU.

70, TITLE:      A Step Towards Interpretable Authorship Verification
http://arxiv.org/abs/2006.12418
AUTHORS:        Oren Halvani ; Lukas Graner ; Roey Regev
COMMENTS:       21 pages, 5 figures. Paper has been accepted for publication in: The 15th International Conference on Availability, Reliability and Security (ARES 2020)
HIGHLIGHT:      To address this problem, we propose an alternative AV approach that considers only topic-agnostic features in its classification decision.

71, TITLE:      Efficient and Phase-aware Video Super-resolution for Cardiac MRI
http://arxiv.org/abs/2005.10626
AUTHORS:        Jhih-Yuan Lin ; Yu-Cheng Chang ; Winston H. Hsu
COMMENTS:       MICCAI 2020
HIGHLIGHT:      To this end, we propose a novel end-to-end trainable network to solve CMR video super-resolution problem without the hardware upgrade and the scanning protocol modifications.

72, TITLE:      srMO-BO-3GP: A sequential regularized multi-objective constrained Bayesian optimization for design applications
http://arxiv.org/abs/2007.03502
AUTHORS:        Anh Tran ; Mike Eldred ; Scott McCann ; Yan Wang
HIGHLIGHT:      In this work, we propose a novel multi-objective (MO) extension, called srMO-BO-3GP, to solve the MO optimization problems in a sequential setting.

73, TITLE:      Distance-Geometric Graph Convolutional Network (DG-GCN)
http://arxiv.org/abs/2007.03513
AUTHORS:        Daniel T. Chang
COMMENTS:       arXiv admin note: substantial text overlap with arXiv:2006.01785
HIGHLIGHT:      To facilitate the incorporation of geometry in deep learning on 3D graphs, we propose a message-passing graph convolutional network based on the distance-geometric graph representation: DG-GCN (distance-geometric graph convolution network).