2020.06.29.txt

==========New Papers==========
1, TITLE:       Duodepth: Static Gesture Recognition Via Dual Depth Sensors
http://arxiv.org/abs/2006.14691
AUTHORS:        Ilya Chugunov ; Avideh Zakhor
COMMENTS:       26th International Conference on Image Processing
HIGHLIGHT:      We present two methodologies for gesture recognition via synchronized recording from two depth cameras to alleviate this occlusion problem.

2, TITLE:       Adaptive additive classification-based loss for deep metric learning
http://arxiv.org/abs/2006.14693
AUTHORS:        Istvan Fehervari ; Ives Macedo
HIGHLIGHT:      In this paper we propose an extension to the existing adaptive margin for classification-based deep metric learning.

3, TITLE:       Learning Data Augmentation with Online Bilevel Optimization for Image Classification
http://arxiv.org/abs/2006.14699
AUTHORS:        Saypraseuth Mounsaveng ; Issam Laradji ; Ismail Ben Ayed ; David Vazquez ; Marco Pedersoli
HIGHLIGHT:      We address this issue by proposing an efficient approach to automatically train a network that learns an effective distribution of transformations to improve its generalization score.

4, TITLE:       ULSAM: Ultra-Lightweight Subspace Attention Module for Compact Convolutional Neural Networks
http://arxiv.org/abs/2006.15102
AUTHORS:        Rajat Saini ; Nandan Kumar Jha ; Bedanta Das ; Sparsh Mittal ; C. Krishna Mohan
COMMENTS:       Accepted as a conference paper in 2020 IEEE Winter Conference on Applications of Computer Vision (WACV)
HIGHLIGHT:      In this work, we propose a simple yet effective "Ultra-Lightweight Subspace Attention Mechanism" (ULSAM), which infers different attention maps for each feature map subspace.

5, TITLE:       Person Re-identification by analyzing Dynamic Variations in Gait Sequences
http://arxiv.org/abs/2006.15109
AUTHORS:        Sandesh Bharadwaj ; Kunal Chanda
COMMENTS:       Presented at ETCCS 2020, accepted for publication in Springer LNEE Proceedings
HIGHLIGHT:      We propose a new approach, which allows recognizing people by analysing the dynamic motion variations and identifying people without using a database of predicted changes.

6, TITLE:       Approximating Euclidean by Imprecise Markov Decision Processes
http://arxiv.org/abs/2006.14923
AUTHORS:        Manfred Jaeger ; Giorgio Bacci ; Giovanni Bacci ; Kim Guldstrand Larsen ; Peter Gjøl Jensen
HIGHLIGHT:      We show that for cost functions over finite time horizons the approximations become arbitrarily precise.

7, TITLE:       Critic Regularized Regression
http://arxiv.org/abs/2006.15134
AUTHORS:        Ziyu Wang ; Alexander Novikov ; Konrad Żołna ; Jost Tobias Springenberg ; Scott Reed ; Bobak Shahriari ; Noah Siegel ; Josh Merel ; Caglar Gulcehre ; Nicolas Heess ; Nando de Freitas
COMMENTS:       23 pages
HIGHLIGHT:      In this paper, we propose a novel offline RL algorithm to learn policies from data using a form of critic-regularized regression (CRR).

8, TITLE:       An Advert Creation System for 3D Product Placements
http://arxiv.org/abs/2006.15131
AUTHORS:        Ivan Bacher ; Hossein Javidnia ; Soumyabrata Dev ; Rahul Agrahari ; Murhaf Hossari ; Matthew Nicholson ; Clare Conran ; Jian Tang ; Peng Song ; David Corrigan ; François Pitié
COMMENTS:       Published in Proc. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2020
HIGHLIGHT:      This paper presents a Video Advertisement Placement & Integration (Adverts) framework, which is capable of perceiving the 3D geometry of the scene and camera motion to blend 3D virtual objects in videos and create the illusion of reality.

9, TITLE:       Deep Q-Network-Driven Catheter Segmentation in 3D US by Hybrid Constrained Semi-Supervised Learning and Dual-UNet
http://arxiv.org/abs/2006.14702
AUTHORS:        Hongxu Yang ; Caifeng Shan ; Alexander F. Kolen ; Peter H. N. de With
COMMENTS:       Accepted by MICCAI 2020
HIGHLIGHT:      In this paper, we propose a novel catheter segmentation approach, which requests fewer annotations than the supervised learning method, but nevertheless achieves better performance.

10, TITLE:      Perspective Plane Program Induction from a Single Image
http://arxiv.org/abs/2006.14708
AUTHORS:        Yikai Li ; Jiayuan Mao ; Xiuming Zhang ; William T. Freeman ; Joshua B. Tenenbaum ; Jiajun Wu
COMMENTS:       CVPR 2020. First two authors contributed equally. Project page: http://p3i.csail.mit.edu/
HIGHLIGHT:      Given an input image, our goal is to induce a neuro-symbolic, program-like representation that jointly models camera poses, object locations, and global scene structures.

11, TITLE:      What they do when in doubt: a study of inductive biases in seq2seq learners
http://arxiv.org/abs/2006.14953
AUTHORS:        Eugene Kharitonov ; Rahma Chaabouni
HIGHLIGHT:      We address that by investigating how popular seq2seq learners generalize in tasks that have high ambiguity in the training data.

12, TITLE:      Investigating and Exploiting Image Resolution for Transfer Learning-based Skin Lesion Classification
http://arxiv.org/abs/2006.14715
AUTHORS:        Amirreza Mahbod ; Gerald Schaefer ; Chunliang Wang ; Rupert Ecker ; Georg Dorffner ; Isabella Ellinger
COMMENTS:       Accepted for the 25th International Conference on Pattern Recognition (ICPR 2020)
HIGHLIGHT:      In this paper, we explore the effect of input image size on skin lesion classification performance of fine-tuned CNNs.

13, TITLE:      CognitiveCNN: Mimicking Human Cognitive Models to resolve Texture-Shape Bias
http://arxiv.org/abs/2006.14722
AUTHORS:        Satyam Mohla ; Anshul Nasery ; Biplab Banerjee ; Subhasis Chaudhari
COMMENTS:       5 Pages; LaTeX; Published at ICLR 2020 Workshop on Bridging AI and Cognitive Science
HIGHLIGHT:      We propose a novel intuitive architecture, namely CognitiveCNN, inspired from feature integration theory in psychology to utilise human-interpretable feature like shape, texture, edges etc. to reconstruct, and classify the image.

14, TITLE:      A Framework for Reinforcement Learning and Planning
http://arxiv.org/abs/2006.15009
AUTHORS:        Thomas M. Moerland ; Joost Broekens ; Catholijn M. Jonker
HIGHLIGHT:      Therefore, this paper presents a unifying framework for reinforcement learning and planning (FRAP), which identifies the underlying dimensions on which any planning or learning algorithm has to decide.

15, TITLE:      Unsupervised Video Decomposition using Spatio-temporal Iterative Inference
http://arxiv.org/abs/2006.14727
AUTHORS:        Polina Zablotskaia ; Edoardo A. Dominici ; Leonid Sigal ; Andreas M. Lehrmann
HIGHLIGHT:      We propose a novel spatio-temporal iterative inference framework that is powerful enough to jointly model complex multi-object representations and explicit temporal dependencies between latent variables across frames.

16, TITLE:      The Fox and the Hound: Comparing Fully Abstract and Robust Compilation
http://arxiv.org/abs/2006.14969
AUTHORS:        Carmine Abate ; Matteo Busi
COMMENTS:       Presented at FCS 2020 workshop
HIGHLIGHT:      We discuss the relation between fully abstract and robust compilation (preservation of satisfaction of arbitrary hyperproperties under adversarial contexts) showing the former implies some variant of the latter.

17, TITLE:      Fast Multi-Level Foreground Estimation
http://arxiv.org/abs/2006.14970
AUTHORS:        Thomas Germer ; Tobias Uelwer ; Stefan Conrad ; Stefan Harmeling
COMMENTS:       Accepted at the 25th International Conference on Pattern Recognition 2020 (ICPR)
HIGHLIGHT:      In this work, we propose a novel method for foreground estimation given the alpha matte.

18, TITLE:      APX-Hardness and Approximation for the k-Burning Number Problem
http://arxiv.org/abs/2006.14733
AUTHORS:        Debajyoti Mondal ; N. Parthiabn ; V. Kavitha ; Indra Rajasingh
HIGHLIGHT:      In this paper we prove that computing $k$-burning number is APX-hard, for any fixed constant $k$.

19, TITLE:      Pre-training via Paraphrasing
http://arxiv.org/abs/2006.15020
AUTHORS:        Mike Lewis ; Marjan Ghazvininejad ; Gargi Ghosh ; Armen Aghajanyan ; Sida Wang ; Luke Zettlemoyer
HIGHLIGHT:      We introduce MARGE, a pre-trained sequence-to-sequence model learned with an unsupervised multi-lingual multi-document paraphrasing objective.

20, TITLE:      Denotational recurrence extraction for amortized analysis
http://arxiv.org/abs/2006.15036
AUTHORS:        Joseph W. Cutler ; Daniel R. Licata ; Norman Danner
COMMENTS:       To appear in ICFP 2020
HIGHLIGHT:      In this paper, we extend these techniques to support amortized analysis, where costs are rearranged from one portion of a program to another to achieve more precise bounds.

21, TITLE:      SAR2SAR: a self-supervised despeckling algorithm for SAR images
http://arxiv.org/abs/2006.15037
AUTHORS:        Emanuele Dalsasso ; Loïc Denis ; Florence Tupin
COMMENTS:       Code is made available at https://github.com/emanueledalsasso/SAR2SAR
HIGHLIGHT:      Due to the difficulty to model the spatial correlation of speckle, a deep learning algorithm with self-supervision is proposed in this paper: SAR2SAR.

22, TITLE:      High Resolution Zero-Shot Domain Adaptation of Synthetically Rendered Face Images
http://arxiv.org/abs/2006.15031
AUTHORS:        Stephan J. Garbin ; Marek Kowalski ; Matthew Johnson ; Jamie Shotton
HIGHLIGHT:      In this work, we propose an algorithm that matches a non-photorealistic, synthetically generated image to a latent vector of a pretrained StyleGAN2 model which, in turn, maps the vector to a photorealistic image of a person of the same pose, expression, hair, and lighting.

23, TITLE:      ProVe -- Self-supervised pipeline for automated product replacement and cold-starting based on neural language models
http://arxiv.org/abs/2006.14994
AUTHORS:        Andrei Ionut Damian ; Laurentiu Piciu ; Cosmin Mihai Marinescu
HIGHLIGHT:      In this paper, we propose a pipeline approach based on Natural Language Understanding, for recommending the most suitable replacements for products that are out-of-stock.

24, TITLE:      An Interactive Data Visualization and Analytics Tool to Evaluate Mobility and Sociability Trends During COVID-19
http://arxiv.org/abs/2006.14882
AUTHORS:        Fan Zuo ; Jingxing Wang ; Jingqin Gao ; Kaan Ozbay ; Xuegang Jeff Ban ; Yubin Shen ; Hong Yang ; Shri Iyer
HIGHLIGHT:      This paper presents the architecture of the COVID related mobility data dashboard and preliminary mobility and sociability metrics for NYC and Seattle.

25, TITLE:      Ensemble Transfer Learning for Emergency Landing Field Identification on Moderate Resource Heterogeneous Kubernetes Cluster
http://arxiv.org/abs/2006.14887
AUTHORS:        Andreas Klos ; Marius Rosenbaum ; Wolfram Schiffmann
HIGHLIGHT:      Thus, based on public available digital orthographic photos and digital surface models, we created various datasets with different sample sizes to facilitate training and testing of neural networks.

26, TITLE:      Determining Image similarity with Quasi-Euclidean Metric
http://arxiv.org/abs/2006.14644
AUTHORS:        Vibhor Singh ; Vishesh Devgan ; Ishu Anand
HIGHLIGHT:      In this paper, we analyzed the similarity between two images from our own novice dataset and assessed its performance against the Euclidean distance metric and SSIM.

27, TITLE:      Constant-Depth and Subcubic-Size Threshold Circuits for Matrix Multiplication
http://arxiv.org/abs/2006.14652
AUTHORS:        Ojas Parekh ; Cynthia A. Phillips ; Conrad D. James ; James B. Aimone
COMMENTS:       Appears in the proceedings of the ACM Symposium on Parallelism in Algorithms and Architectures (SPAA), 2018
HIGHLIGHT:      We describe a theoretical approach for multiplying two $N$ by $N$ matrices that integrates threshold gate logic with conventional fast matrix multiplication algorithms, that perform $O(N^\omega)$ arithmetic operations for a positive constant $\omega < 3$.

28, TITLE:      Biologically Plausible Learning of Text Representation with Spiking Neural Networks
http://arxiv.org/abs/2006.14894
AUTHORS:        Marcin Białas ; Marcin Michał Mirończuk ; Jacek Mańdziuk
COMMENTS:       This article was originally submitted for Parallel Problem Solving from Nature conference and will be available in Springer Lecture Notes in Computer Science (LNCS)
HIGHLIGHT:      This study proposes a novel biologically plausible mechanism for generating low-dimensional spike-based text representation.

29, TITLE:      Can 3D Adversarial Logos Cloak Humans?
http://arxiv.org/abs/2006.14655
AUTHORS:        Tianlong Chen ; Yi Wang ; Jingyang Zhou ; Sijia Liu ; Shiyu Chang ; Chandrajit Bajaj ; Zhangyang Wang
HIGHLIGHT:      This paper presents a new 3D adversarial logo attack: we construct an arbitrary shape logo from a 2D texture image and map this image into a 3D adversarial logo via a texture mapping called logo transformation.

30, TITLE:      Cascaded Convolutional Neural Networks with Perceptual Loss for Low Dose CT Denoising
http://arxiv.org/abs/2006.14738
AUTHORS:        Sepehr Ataei ; Dr. Javad Alirezaie ; Dr. Paul Babyn
HIGHLIGHT:      Low Dose CT Denoising research aims to reduce the risks of radiation exposure to patients.

31, TITLE:      Graph Optimal Transport for Cross-Domain Alignment
http://arxiv.org/abs/2006.14744
AUTHORS:        Liqun Chen ; Zhe Gan ; Yu Cheng ; Linjie Li ; Lawrence Carin ; Jingjing Liu
HIGHLIGHT:      We propose Graph Optimal Transport (GOT), a principled framework that germinates from recent advances in Optimal Transport (OT).

32, TITLE:      The State of AI Ethics Report (June 2020)
http://arxiv.org/abs/2006.14662
AUTHORS:        Abhishek Gupta ; Camylle Lanteigne ; Victoria Heath ; Marianna Bergamaschi Ganapini ; Erick Galinkin ; Allison Cohen ; Tania De Gasperis ; Mo Akif ; Renjie Butalid
COMMENTS:       128 pages
HIGHLIGHT:      We cover a wide set of areas in this report spanning Agency and Responsibility, Security and Risk, Disinformation, Jobs and Labor, the Future of AI Ethics, and more.

33, TITLE:      SPSG: Self-Supervised Photometric Scene Generation from RGB-D Scans
http://arxiv.org/abs/2006.14660
AUTHORS:        Angela Dai ; Yawar Siddiqui ; Justus Thies ; Julien Valentin ; Matthias Nießner
COMMENTS:       Video: https://youtu.be/1cj962m9zqo
HIGHLIGHT:      We present SPSG, a novel approach to generate high-quality, colored 3D models of scenes from RGB-D scan observations by learning to infer unobserved scene geometry and color in a self-supervised fashion.

34, TITLE:      Suggestive Annotation of Brain Tumour Images with Gradient-guided Sampling
http://arxiv.org/abs/2006.14984
AUTHORS:        Chengliang Dai ; Shuo Wang ; Yuanhan Mo ; Elsa Angelini ; Yike Guo ; Wenjia Bai
COMMENTS:       Paper accepted by MICCAI 2020
HIGHLIGHT:      In this paper, we propose an efficient annotation framework for brain tumour images that is able to suggest informative sample images for human experts to annotate.

35, TITLE:      LPar -- A Distributed Multi Agent platform for building Polyglot, Omni Channel and Industrial grade Natural Language Interfaces
http://arxiv.org/abs/2006.14666
AUTHORS:        Pranav Sharma
HIGHLIGHT:      To address these challenges, we introduce LPar, a distributed multi agent platform for large scale industrial deployment of polyglot, diverse and inter-operable agents.

36, TITLE:      THEaiTRE: Artificial Intelligence to Write a Theatre Play
http://arxiv.org/abs/2006.14668
AUTHORS:        Rudolf Rosa ; Ondřej Dušek ; Tom Kocmi ; David Mareček ; Tomáš Musil ; Patrícia Schmidtová ; Dominik Jurko ; Ondřej Bojar ; Daniel Hrbek ; David Košťák ; Martina Kinská ; Josef Doležal ; Klára Vosecká
COMMENTS:       accepted to AI4Narratives2020
HIGHLIGHT:      We present THEaiTRE, a starting project aimed at automatic generation of theatre play scripts.

37, TITLE:      Point Proposal Network for Reconstructing 3D Particle Positions with Sub-Pixel Precision in Liquid Argon Time Projection Chambers
http://arxiv.org/abs/2006.14745
AUTHORS:        Laura Dominé ; Kazuhiro Terao
HIGHLIGHT:      Using the PILArNet public LArTPC data sample as a benchmark, our algorithm successfully predicted 96.8%, 97.8%, and 98.1% of 3D points within the voxel distance of 3, 10, and 20 from the provided true point locations respectively.

38, TITLE:      Meta Deformation Network: Meta Functionals for Shape Correspondence
http://arxiv.org/abs/2006.14758
AUTHORS:        Daohan Lu ; Yi Fang
HIGHLIGHT:      We present a new technique named "Meta Deformation Network" for 3D shape matching via deformation, in which a deep neural network maps a reference shape onto the parameters of a second neural network whose task is to give the correspondence between a learned template and query shape via deformation.

39, TITLE:      Deepfake Detection using Spatiotemporal Convolutional Networks
http://arxiv.org/abs/2006.14749
AUTHORS:        Oscar de Lima ; Sean Franklin ; Shreshtha Basu ; Blake Karwoski ; Annet George
HIGHLIGHT:      Better generative models and larger datasets have led to more realistic fake videos that can fool the human eye but produce temporal and spatial artifacts that deep learning approaches can detect. We created a benchmark of the performance of spatiotemporal convolutional methods using the Celeb-DF dataset.

40, TITLE:      Fully Convolutional Open Set Segmentation
http://arxiv.org/abs/2006.14673
AUTHORS:        Hugo Oliveira ; Caio Silva ; Gabriel L. S. Machado ; Keiller Nogueira ; Jefersson A. dos Santos
COMMENTS:       Submitted to the Machine Learning Journal
HIGHLIGHT:      In this paper, we discuss the limitations of Closed Set segmentation and propose two fully convolutional approaches to effectively address Open Set semantic segmentation: OpenFCN and OpenPCS.

41, TITLE:      Lesion Mask-based Simultaneous Synthesis of Anatomic and MolecularMR Images using a GAN
http://arxiv.org/abs/2006.14761
AUTHORS:        Pengfei Guo ; Puyang Wang ; Jinyuan Zhou ; Vishal Patel ; Shanshan Jiang
COMMENTS:       MICCAI 2020
HIGHLIGHT:      In this paper, we propose a generative adversarial network (GAN), which can simultaneously synthesize data from arbitrary manipulated lesion information on multiple anatomic and molecular MRI sequences, including T1-weighted (T1w), gadolinium enhanced T1w (Gd-T1w), T2-weighted (T2w), fluid-attenuated inversion recovery (FLAIR), and amide proton transfer-weighted (APTw).

42, TITLE:      Supermasks in Superposition
http://arxiv.org/abs/2006.14769
AUTHORS:        Mitchell Wortsman ; Vivek Ramanujan ; Rosanne Liu ; Aniruddha Kembhavi ; Mohammad Rastegari ; Jason Yosinski ; Ali Farhadi
HIGHLIGHT:      We present the Supermasks in Superposition (SupSup) model, capable of sequentially learning thousands of tasks without catastrophic forgetting.

43, TITLE:      Dialog as a Vehicle for Lifelong Learning
http://arxiv.org/abs/2006.14767
AUTHORS:        Aishwarya Padmakumar ; Raymond J. Mooney
COMMENTS:       Position Paper Track at the SIGDIAL Special Session on Physically Situated Dialogue (RoboDial 2.0) - Camera Ready Version
HIGHLIGHT:      In this position paper, we present the problem of designing dialog systems that enable lifelong learning as an important challenge problem, in particular for applications involving physically situated robots.

44, TITLE:      Pushing the Limit of Unsupervised Learning for Ultrasound Image Artifact Removal
http://arxiv.org/abs/2006.14773
AUTHORS:        Shujaat Khan ; Jaeyoung Huh ; Jong Chul Ye
HIGHLIGHT:      In this paper, inspired by the recent theory of unsupervised learning using optimal transport driven cycleGAN (OT-cycleGAN), we investigate applicability of unsupervised deep learning for US artifact removal problems without matched reference data.

45, TITLE:      Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance
http://arxiv.org/abs/2006.14779
AUTHORS:        Gagan Bansal ; Tongshuang Wu ; Joyce Zhu ; Raymond Fok ; Besmira Nushi ; Ece Kamar ; Marco Tulio Ribeiro ; Daniel S. Weld
COMMENTS:       Draft/pre-print
HIGHLIGHT:      Can we develop explanatory approaches that help humans decide whether and when to trust AI input?

46, TITLE:      Blind Image Deconvolution using Student's-t Prior with Overlapping Group Sparsity
http://arxiv.org/abs/2006.14780
AUTHORS:        In S. Jeon ; Deokyoung Kang ; Suk I. Yoo
HIGHLIGHT:      In this paper, we solve blind image deconvolution problem that is to remove blurs form a signal degraded image without any knowledge of the blur kernel.

47, TITLE:      Ricci Curvature Based Volumetric Segmentation of the Auditory Ossicles
http://arxiv.org/abs/2006.14788
AUTHORS:        Na Lei ; Jisui Huang ; Yuxue Ren ; Emil Saucan ; Zhenchang Wang
HIGHLIGHT:      We therefore propose a completely automatic method to locate the ossicles which requires neither templates, nor manual labels.

48, TITLE:      Unsupervised Discovery of Object Landmarks via Contrastive Learning
http://arxiv.org/abs/2006.14787
AUTHORS:        Zezhou Cheng ; Jong-Chyi Su ; Subhransu Maji
COMMENTS:       Project Page: https://people.cs.umass.edu/~zezhoucheng/contrastive_landmark Code: https://github.com/cvl-umass/ContrastLandmark
HIGHLIGHT:      In this paper, we develop a simple and effective approach based on contrastive learning of invariant representations.

49, TITLE:      Quantum Communication Complexity of Distribution Testing
http://arxiv.org/abs/2006.14870
AUTHORS:        Aleksandrs Belovs ; Arturo Castellanos ; François Le Gall ; Guillaume Malod ; Alexander A. Sherstov
COMMENTS:       11 pages
HIGHLIGHT:      In the present paper we show that the quantum communication complexity of this problem is $\tilde{O}(n/(t\epsilon^2))$ qubits when the distributions have low $l_2$-norm, which gives a quadratic improvement over the classical communication complexity obtained by Andoni, Malkin and Nosatzki.

50, TITLE:      TURL: Table Understanding through Representation Learning
http://arxiv.org/abs/2006.14806
AUTHORS:        Xiang Deng ; Huan Sun ; Alyssa Lees ; You Wu ; Cong Yu
COMMENTS:       Our source code, benchmark, as well as pre-trained models will be available on https://github.com/sunlab-osu/TURL
HIGHLIGHT:      In this paper, we present TURL, a novel framework that introduces the pre-training/finetuning paradigm to relational Web tables.

51, TITLE:      Explanation Augmented Feedback in Human-in-the-Loop Reinforcement Learning
http://arxiv.org/abs/2006.14804
AUTHORS:        Lin Guan ; Mudit Verma ; Subbarao Kambhampati
HIGHLIGHT:      To address this, we present EXPAND (Explanation Augmented Feedback) which allows for explanatory information to be given as saliency maps from the human in addition to the binary feedback.

52, TITLE:      Text Detection on Roughly Placed Books by Leveraging a Learning-based Model Trained with Another Domain Data
http://arxiv.org/abs/2006.14808
AUTHORS:        Riku Anegawa ; Masayoshi Aritsugi
HIGHLIGHT:      In this paper, we focus on how to generate bounding boxes that are appropriate to grasp text areas on books to help implement automatic text detection.

53, TITLE:      Cross-Ssupervised Object Detection
http://arxiv.org/abs/2006.15056
AUTHORS:        Zitian Chen ; Zhiqiang Shen ; Jiahui Yu ; Erik Learned-Miller
HIGHLIGHT:      In this work, we show how to build better object detectors from weakly labeled images of new categories by leveraging knowledge learned from fully labeled base categories.

54, TITLE:      A Loss Function for Generative Neural Networks Based on Watson's Perceptual Model
http://arxiv.org/abs/2006.15057
AUTHORS:        Steffen Czolbe ; Oswin Krause ; Ingemar Cox ; Christian Igel
HIGHLIGHT:      We propose such a loss function based on Watson's perceptual model, which computes a weighted distance in frequency space and accounts for luminance and contrast masking.

55, TITLE:      Computing Light Transport Gradients using the Adjoint Method
http://arxiv.org/abs/2006.15059
AUTHORS:        Jos Stam
COMMENTS:       23 pages, 8 figures, unpublished manuscript
HIGHLIGHT:      The key insight of this paper is that computing gradients in Transport Theory is akin to computing the importance, a quantity adjoint to radiance that satisfies an adjoint equation.

56, TITLE:      Few-Shot Anomaly Detection for Polyp Frames from Colonoscopy
http://arxiv.org/abs/2006.14811
AUTHORS:        Yu Tian ; Gabriel Maicas ; Leonardo Zorron Cheng Tao Pu ; Rajvinder Singh ; Johan W. Verjans ; Gustavo Carneiro
COMMENTS:       Accept at MICCAI 2020
HIGHLIGHT:      In this paper, we address the inappropriate sensitivity to outliers by also learning from inliers.

57, TITLE:      Evaluation of Text Generation: A Survey
http://arxiv.org/abs/2006.14799
AUTHORS:        Asli Celikyilmaz ; Elizabeth Clark ; Jianfeng Gao
COMMENTS:       42 pages
HIGHLIGHT:      For each category, we discuss the progress that has been made and the challenges still being faced, with a focus on the evaluation of recently proposed NLG tasks and neural NLG models.

58, TITLE:      Training Convolutional ReLU Neural Networks in Polynomial Time: Exact Convex Optimization Formulations
http://arxiv.org/abs/2006.14798
AUTHORS:        Tolga Ergen ; Mert Pilanci
HIGHLIGHT:      We study training of Convolutional Neural Networks (CNNs) with ReLU activations and introduce exact convex optimization formulations with a polynomial complexity with respect to the number of data samples, the number of neurons and data dimension.

59, TITLE:      Object-Centric Learning with Slot Attention
http://arxiv.org/abs/2006.15055
AUTHORS:        Francesco Locatello ; Dirk Weissenborn ; Thomas Unterthiner ; Aravindh Mahendran ; Georg Heigold ; Jakob Uszkoreit ; Alexey Dosovitskiy ; Thomas Kipf
HIGHLIGHT:      In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots.

60, TITLE:      AvE: Assistance via Empowerment
http://arxiv.org/abs/2006.14796
AUTHORS:        Yuqing Du ; Stas Tiomkin ; Emre Kiciman ; Daniel Polani ; Pieter Abbeel ; Anca Dragan
HIGHLIGHT:      We propose a new paradigm for assistance by instead increasing the human's ability to control their environment, and formalize this approach by augmenting reinforcement learning with human empowerment.

61, TITLE:      Graph modelling approaches for motorway traffic flow prediction
http://arxiv.org/abs/2006.14824
AUTHORS:        Adriana-Simona Mihaita ; Zac Papachatgis ; Marian-Andrei Rizoiu
HIGHLIGHT:      This paper presents two new spatial-temporal approaches for building accurate short-term prediction along a popular motorway in Sydney, by making use of the graph structure of the motorway network (including exits and entries).

62, TITLE:      A survey of loss functions for semantic segmentation
http://arxiv.org/abs/2006.14822
AUTHORS:        Shruti Jadon
COMMENTS:       5 pages, 5 figures, 2 tables
HIGHLIGHT:      In this paper, we have summarized most of the well-known loss functions widely used in Image segmentation and listed out the cases where their usage can help in fast and better convergence of a Model.

63, TITLE:      Intrinsic Reward Driven Imitation Learning via Generative Model
http://arxiv.org/abs/2006.15061
AUTHORS:        Xingrui Yu ; Yueming Lyu ; Ivor W. Tsang
HIGHLIGHT:      To address this challenge, we propose a novel reward learning module to generate intrinsic reward signals via a generative model.

64, TITLE:      Lee-Yang zeros and the complexity of the ferromagnetic Ising Model on bounded-degree graphs
http://arxiv.org/abs/2006.14828
AUTHORS:        Pjotr Buy ; Andreas Galanis ; Viresh Patel ; Guus Regts
COMMENTS:       38 pages, 1 figure
HIGHLIGHT:      We study the computational complexity of approximating the partition function of the ferromagnetic Ising model in the Lee-Yang circle of zeros given by $|\lambda|=1$, where $\lambda$ is the external field of the model.

65, TITLE:      Continual Learning from the Perspective of Compression
http://arxiv.org/abs/2006.15078
AUTHORS:        Xu He ; Min Lin
COMMENTS:       4th Lifelong Learning Workshop at ICML 2020
HIGHLIGHT:      To address these limitations, we propose a new continual learning method that combines ML plug-in and Bayesian mixture codes.

66, TITLE:      Expandable YOLO: 3D Object Detection from RGB-D Images
http://arxiv.org/abs/2006.14837
AUTHORS:        Masahiro Takahashi ; Alessandro Moro ; Yonghoon Ji ; Kazunori Umeda
COMMENTS:       5 pages, 8 figures
HIGHLIGHT:      This paper aims at constructing a light-weight object detector that inputs a depth and a color image from a stereo camera.

67, TITLE:      Not all Failure Modes are Created Equal: Training Deep Neural Networks for Explicable (Mis)Classification
http://arxiv.org/abs/2006.14841
AUTHORS:        Alberto Olmo ; Sailik Sengupta ; Subbarao Kambhampati
COMMENTS:       The first two authors contributed equally
HIGHLIGHT:      Thus, in this work, we aim to reduce inexplicable errors.

68, TITLE:      What can I do here? A Theory of Affordances in Reinforcement Learning
http://arxiv.org/abs/2006.15085
AUTHORS:        Khimya Khetarpal ; Zafarali Ahmed ; Gheorghe Comanici ; David Abel ; Doina Precup
COMMENTS:       Thirty-seventh International Conference on Machine Learning (ICML 2020)
HIGHLIGHT:      In this paper, we develop a theory of affordances for agents who learn and plan in Markov Decision Processes.

69, TITLE:      End-to-end training of deep kernel map networks for image classification
http://arxiv.org/abs/2006.15088
AUTHORS:        Mingyuan Jiu ; Hichem Sahbi
HIGHLIGHT:      In this paper, we introduce a novel "end-to-end" design for deep kernel map learning that balances the approximation quality of kernels and their discrimination power.

70, TITLE:      An Automatic Reader of Identity Documents
http://arxiv.org/abs/2006.14853
AUTHORS:        Filippo Attivissimo ; Nicola Giaquinto ; Marco Scarpetta ; Maurizio Spadavecchia
COMMENTS:       6 pages, 9 figures
HIGHLIGHT:      In this paper the prototype of a novel automatic reading system of identity documents is presented.

71, TITLE:      AutoSNAP: Automatically Learning Neural Architectures for Instrument Pose Estimation
http://arxiv.org/abs/2006.14858
AUTHORS:        David Kügler ; Marc Uecker ; Arjan Kuijper ; Anirban Mukhopadhyay
COMMENTS:       Accepted at MICCAI 2020 Preparing code for release at https://github.com/MECLabTUDA/AutoSNAP
HIGHLIGHT:      We propose an automatic framework (AutoSNAP) for instrument pose estimation problems, which discovers and learns the architectures for neural networks.

72, TITLE:      Orthogonal Deep Models As Defense Against Black-Box Attacks
http://arxiv.org/abs/2006.14856
AUTHORS:        Mohammad A. A. K. Jalwana ; Naveed Akhtar ; Mohammed Bennamoun ; Ajmal Mian
COMMENTS:       Accepted in IEEE Access
HIGHLIGHT:      Detailed empirical study verifies that controlled misalignment of gradients under our orthogonality objective significantly boosts a model's robustness against transferable black-box adversarial attacks.

73, TITLE:      Designing and Learning Trainable Priors with Non-Cooperative Games
http://arxiv.org/abs/2006.14859
AUTHORS:        Bruno Lecouat ; Jean Ponce ; Julien Mairal
HIGHLIGHT:      We introduce a general framework for designing and learning neural networks whose forward passes can be interpreted as solving convex optimization problems, and whose architectures are derived from an optimization algorithm.

74, TITLE:      RPM-Net: Recurrent Prediction of Motion and Parts from Point Cloud
http://arxiv.org/abs/2006.14865
AUTHORS:        Zihao Yan ; Ruizhen Hu ; Xingguang Yan ; Luanmin Chen ; Oliver van Kaick ; Hao Zhang ; Hui Huang
COMMENTS:       Accepted to SIGGRAPH Asia 2019, project page at https://vcc.tech/research/2019/RPMNet
HIGHLIGHT:      We introduce RPM-Net, a deep learning-based approach which simultaneously infers movable parts and hallucinates their motions from a single, un-segmented, and possibly partial, 3D point cloud shape.

75, TITLE:      Domain Contrast for Domain Adaptive Object Detection
http://arxiv.org/abs/2006.14863
AUTHORS:        Feng Liu ; Xiaoxong Zhang ; Fang Wan ; Xiangyang Ji ; Qixiang Ye
HIGHLIGHT:      We present Domain Contrast (DC), a simple yet effective approach inspired by contrastive learning for training domain adaptive detectors.



==========Updates to Previous Papers==========
1, TITLE:       A Recent Survey on the Applications of Genetic Programming in Image Processing
http://arxiv.org/abs/1901.07387
AUTHORS:        Asifullah Khan ; Aqsa Saeed Qureshi ; Noorul Wahab ; Mutawara Hussain ; Muhammad Yousaf Hamza
COMMENTS:       31 pages, 12 figures, and 1 table
HIGHLIGHT:      This survey thus presents the diverse applications of GP in Image Processing and provides useful resources for further research.

2, TITLE:       Deep-Learning Inversion of Seismic Data
http://arxiv.org/abs/1901.07733
AUTHORS:        Shucai Li ; Bin Liu ; Yuxiao Ren ; Yangkang Chen ; Senlin Yang ; Yunhai Wang ; Peng Jiang
HIGHLIGHT:      We propose a new method to tackle the mapping challenge from time-series data to spatial image in the field of seismic exploration, i.e., reconstructing the velocity model directly from seismic data by deep neural networks (DNNs).

3, TITLE:       Riccati-based feedback stabilization for unstable Power system models
http://arxiv.org/abs/2006.14210
AUTHORS:        Mahtab Uddin ; M. Monir Uddin ; Md. Abdul Hakim Khan
COMMENTS:       28 pages, 19 figures
HIGHLIGHT:      In this article, the objective is mainly focused on finding optimal control for the large-scale sparse unstable power system models using optimal feedback matrix achieved by the Riccati-based feedback stabilization process.

4, TITLE:       Deep Residual 3D U-Net for Joint Segmentation and Texture Classification of Nodules in Lung
http://arxiv.org/abs/2006.14215
AUTHORS:        Alexandr G. Rassadin
COMMENTS:       10 pages, 5 figures, 2 tables, accepted for publication at ICIAR 2020 (LNDb Grand Challenge)
HIGHLIGHT:      In this work we present a method for lung nodules segmentation, their texture classification and subsequent follow-up recommendation from the CT image of lung.

5, TITLE:       Derandomization from Algebraic Hardness
http://arxiv.org/abs/1905.00091
AUTHORS:        Zeyu Guo ; Mrinal Kumar ; Ramprasad Saptharishi ; Noam Solomon
COMMENTS:       Incorporated some reviewer comments, extension of the main theorems to HSGs from k-variate polynomials for small-enough k, connection to the tau-conjecture of Shub-Smale
HIGHLIGHT:      In this paper, we give a new construction of such an HSG assuming that we have an explicit polynomial of sufficient hardness.

6, TITLE:       IIT Gandhinagar at SemEval-2020 Task 9: Code-Mixed Sentiment Classification Using Candidate Sentence Generation and Selection
http://arxiv.org/abs/2006.14465
AUTHORS:        Vivek Srivastava ; Mayank Singh
HIGHLIGHT:      We present a candidate sentence generation and selection based approach on top of the Bi-LSTM based neural classifier to classify the Hinglish code-mixed text into one of the three sentiment classes positive, negative, or neutral.

7, TITLE:       On Leveraging Pretrained GANs for Generation with Limited Data
http://arxiv.org/abs/2002.11810
AUTHORS:        Miaoyun Zhao ; Yulai Cong ; Lawrence Carin
COMMENTS:       Accepted at ICML 2020
HIGHLIGHT:      To further adapt the transferred filters to the target domain, we propose adaptive filter modulation (AdaFM).

8, TITLE:       VisuoSpatial Foresight for Multi-Step, Multi-Task Fabric Manipulation
http://arxiv.org/abs/2003.09044
AUTHORS:        Ryan Hoque ; Daniel Seita ; Ashwin Balakrishna ; Aditya Ganapathi ; Ajay Kumar Tanwani ; Nawid Jamali ; Katsu Yamane ; Soshi Iba ; Ken Goldberg
COMMENTS:       Robotics: Science and Systems (RSS) 2020
HIGHLIGHT:      We introduce VisuoSpatial Foresight (VSF), which builds on prior work by learning visual dynamics on domain randomized RGB images and depth maps simultaneously and completely in simulation.

9, TITLE:       Differentiable Product Quantization for End-to-End Embedding Compression
http://arxiv.org/abs/1908.09756
AUTHORS:        Ting Chen ; Lala Li ; Yizhou Sun
COMMENTS:       ICML'2020. Code at https://github.com/chentingpc/dpq_embedding_compression
HIGHLIGHT:      In this work, we propose a generic and end-to-end learnable compression framework termed differentiable product quantization (DPQ).

10, TITLE:      A Broader Study of Cross-Domain Few-Shot Learning
http://arxiv.org/abs/1912.07200
AUTHORS:        Yunhui Guo ; Noel C. Codella ; Leonid Karlinsky ; James V. Codella ; John R. Smith ; Kate Saenko ; Tajana Rosing ; Rogerio Feris
HIGHLIGHT:      In this paper, we propose the Broader Study of Cross-Domain Few-Shot Learning (BSCD-FSL) benchmark, consisting of image data from a diverse assortment of image acquisition methods.

11, TITLE:      Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation
http://arxiv.org/abs/2002.08546
AUTHORS:        Jian Liang ; Dapeng Hu ; Jiashi Feng
COMMENTS:       ICML 2020 camera ready. Code is available at https://github.com/tim-learn/SHOT
HIGHLIGHT:      We propose a simple yet generic representation learning framework, named \emph{Source HypOthesis Transfer} (SHOT).

12, TITLE:      Stochastic Online Optimization using Kalman Recursion
http://arxiv.org/abs/2002.03636
AUTHORS:        Joseph de Vilmarest ; Olivier Wintenberger
HIGHLIGHT:      In order to avoid any projection step we propose a two-phase analysis.

13, TITLE:      Rethinking Curriculum Learning with Incremental Labels and Adaptive Compensation
http://arxiv.org/abs/2001.04529
AUTHORS:        Madan Ravi Ganesh ; Jason J. Corso
COMMENTS:       8 pages
HIGHLIGHT:      In this work, we propose Learning with Incremental Labels and Adaptive Compensation(LILAC), which takes a novel approach to curriculum learning.

14, TITLE:      LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery
http://arxiv.org/abs/1905.02744
AUTHORS:        Junming Zhang ; Manikandasriram Srinivasan Ramanagopal ; Ram Vasudevan ; Matthew Johnson-Roberson
COMMENTS:       14 pages, 3 figures, 5 tables
HIGHLIGHT:      This paper combines these approaches together to generate high-quality dense depth maps.

15, TITLE:      Resisting Large Data Variations via Introspective Transformation Network
http://arxiv.org/abs/1805.06447
AUTHORS:        Yunhan Zhao ; Ye Tian ; Charless Fowlkes ; Wei Shen ; Alan Yuille
COMMENTS:       camera-ready version, WACV 2020
HIGHLIGHT:      In this paper, we propose a principled approach to train networks with significantly improved resistance to large variations between training and testing data.

16, TITLE:      Hierarchical Character Embeddings: Learning Phonological and Semantic Representations in Languages of Logographic Origin using Recursive Neural Networks
http://arxiv.org/abs/1912.09913
AUTHORS:        Minh Nguyen ; Gia H. Ngo ; Nancy F. Chen
COMMENTS:       Accepted by IEEE Transactions on Audio, Speech and Language Processing. Copyright 2019 IEEE
HIGHLIGHT:      We propose building hierarchical logograph (character) embeddings from logograph recursive structures using treeLSTM, a recursive neural network.

17, TITLE:      On equivalence, languages equivalence and minimization of multi-letter and multi-letter measure-many quantum automata
http://arxiv.org/abs/1203.0113
AUTHORS:        Tianrong Lin
COMMENTS:       30 pages, conclusion section corrected
HIGHLIGHT:      The direct consequences of the above outcomes are summarized in the paper.

18, TITLE:      Entropy Minimization In Emergent Languages
http://arxiv.org/abs/1905.13687
AUTHORS:        Eugene Kharitonov ; Rahma Chaabouni ; Diane Bouchacourt ; Marco Baroni
COMMENTS:       Accepted at ICML 2020
HIGHLIGHT:      We investigate here the information-theoretic complexity of such languages, focusing on the basic two-agent, one-exchange setup.

19, TITLE:      Deep Learning Inversion of Electrical Resistivity Data
http://arxiv.org/abs/1904.05265
AUTHORS:        Bin Liu ; Qian Guo ; Shucai Li ; Benchao Liu ; Yuxiao Ren ; Yonghao Pang ; Xu Guo ; Lanbo Liu ; Peng Jiang
COMMENTS:       IEEE Transactions on Geoscience and Remote Sensing, 2020
HIGHLIGHT:      Inspired by the remarkable nonlinear mapping ability of deep learning approaches, in this article, we propose to build the mapping from apparent resistivity data (input) to resistivity model (output) directly by convolutional neural networks (CNNs).

20, TITLE:      Slimming Neural Networks using Adaptive Connectivity Scores
http://arxiv.org/abs/2006.12463
AUTHORS:        Madan Ravi Ganesh ; Dawsin Blanchard ; Jason J. Corso ; Salimeh Yasaei Sekeh
COMMENTS:       22 pages
HIGHLIGHT:      In this work,we propose Slimming Neural networks using Adaptive Connectivity Measures(SNACS), as an algorithm that uses a probabilistic framework for compression while incorporating weight-based constraints at multiple levels to capitalize on both their strengths and overcome previous issues. To reduce the amount of unnecessary manual effort required to set the upper pruning limit of different layers in a DNN we propose a set of operating constraints to help automatically set them.

21, TITLE:      Optimizing AI for Teamwork
http://arxiv.org/abs/2004.13102
AUTHORS:        Gagan Bansal ; Besmira Nushi ; Ece Kamar ; Eric Horvitz ; Daniel S. Weld
COMMENTS:       Pre-print/Draft
HIGHLIGHT:      So, we propose training AI systems in a human-centered manner and directly optimizing for team performance.

22, TITLE:      Towards an Effective and Efficient Deep Learning Model for COVID-19 Patterns Detection in X-ray Images
http://arxiv.org/abs/2004.05717
AUTHORS:        Eduardo Luz ; Pedro Lopes Silva ; Rodrigo Silva ; Ludmila Silva ; Gladston Moreira ; David Menotti
COMMENTS:       31 pages, 9 figures
HIGHLIGHT:      Thus, the main goal of this work is to propose an accurate yet efficient method in terms of memory and processing time for the problem of COVID-19 screening in chest X-rays.

23, TITLE:      A Novel and Reliable Deep Learning Web-Based Tool to Detect COVID-19 Infection from Chest CT-Scan
http://arxiv.org/abs/2006.14419
AUTHORS:        Abdolkarim Saeedi ; Maryam Saeedi ; Arash Maghsoudi
COMMENTS:       9 pages, 8 figures, Improved English writing in the abstract, introduction and conclusion sections. Removed DenseNet typos underneath figures. Fixed some other minor typos
HIGHLIGHT:      In this work, we introduce a computer aided diagnosis (CAD) web service to detect COVID- 19 online.

24, TITLE:      PraNet: Parallel Reverse Attention Network for Polyp Segmentation
http://arxiv.org/abs/2006.11392
AUTHORS:        Deng-Ping Fan ; Ge-Peng Ji ; Tao Zhou ; Geng Chen ; Huazhu Fu ; Jianbing Shen ; Ling Shao
COMMENTS:       Accepted to MICCAI 2020
HIGHLIGHT:      To address these challenges, we propose a parallel reverse attention network (PraNet) for accurate polyp segmentation in colonoscopy images.

25, TITLE:      Superpixel Segmentation via Convolutional Neural Networks with Regularized Information Maximization
http://arxiv.org/abs/2002.06765
AUTHORS:        Teppei Suzuki
COMMENTS:       To appear in ICASSP 2020
HIGHLIGHT:      We propose an unsupervised superpixel segmentation method by optimizing a randomly-initialized convolutional neural network (CNN) in inference time.

26, TITLE:      Deep Residual Network based food recognition for enhanced Augmented Reality application
http://arxiv.org/abs/2005.04292
AUTHORS:        Siddarth S ; Sainath G ; Vignesh S
COMMENTS:       Total Pages:7 Total Figures:10
HIGHLIGHT:      Deep neural network based learning approaches is widely utilized for image classification or object detection based problems with remarkable outcomes.

27, TITLE:      A Self-Attentional Neural Architecture for Code Completion with Multi-Task Learning
http://arxiv.org/abs/1909.06983
AUTHORS:        Fang Liu ; Ge Li ; Bolin Wei ; Xin Xia ; Zhiyi Fu ; Zhi Jin
COMMENTS:       Accepted by International Conference on Program Comprehension (ICPC 2020)
HIGHLIGHT:      To utilize the hierarchical structural information of the programs, we present a novel method that considers the path from the predicting node to the root node.

28, TITLE:      Combining Deep Learning and Verification for Precise Object Instance Detection
http://arxiv.org/abs/1912.12270
AUTHORS:        Siddharth Ancha ; Junyu Nan ; David Held
COMMENTS:       9 pages main paper, 2 pages references, 10 pages supplementary material
HIGHLIGHT:      Based on an approximation to this framework, we present a practical detection system that can verify, with high precision, whether each detection of a machine-learning based object detector is correct. To achieve this, we have developed a set of verification tests which a proposed detection must pass to be accepted.

29, TITLE:      Fully Trainable and Interpretable Non-Local Sparse Models for Image Restoration
http://arxiv.org/abs/1912.02456
AUTHORS:        Bruno Lecouat ; Jean Ponce ; Julien Mairal
HIGHLIGHT:      We propose a novel differentiable relaxation of joint sparsity that exploits both principles and leads to a general framework for image restoration which is (1) trainable end to end, (2) fully interpretable, and (3) much more compact than competing deep learning architectures.

30, TITLE:      Resisting Crowd Occlusion and Hard Negatives for Pedestrian Detection in the Wild
http://arxiv.org/abs/2005.07344
AUTHORS:        Zhe Wang ; Jun Wang ; Yezhou Yang
COMMENTS:       10 pages, 6 figures
HIGHLIGHT:      In this paper, we offer two approaches based on the general region-based detection framework to tackle these challenges.

31, TITLE:      Digital Social Contracts: A Foundation for an Egalitarian and Just Digital Society
http://arxiv.org/abs/2005.06261
AUTHORS:        Luca Cardelli ; Liav Orgad ; Gal Shahaf ; Ehud Shapiro ; Nimrod Talmon
HIGHLIGHT:      Here, we present a formal definition of a digital social contract as agents that communicate asynchronously via crypto-speech acts, where the output of each agent is the input of all the other agents.

32, TITLE:      A CNN-LSTM Architecture for Detection of Intracranial Hemorrhage on CT scans
http://arxiv.org/abs/2005.10992
AUTHORS:        Nhan T. Nguyen ; Dat Q. Tran ; Nghia T. Nguyen ; Ha Q. Nguyen
HIGHLIGHT:      We propose a novel method that combines a convolutional neural network (CNN) with a long short-term memory (LSTM) mechanism for accurate prediction of intracranial hemorrhage on computed tomography (CT) scans.

33, TITLE:      A Solution of the P versus NP Problem based on specific property of clique function
http://arxiv.org/abs/1911.00722
AUTHORS:        Boyu Sima
HIGHLIGHT:      Razborov has proved superpolynomial lower bounds for monotone circuits by using method of approximation.

34, TITLE:      An Implicit Attention Mechanism for Deep Learning Pedestrian Re-identification Frameworks
http://arxiv.org/abs/2001.11267
AUTHORS:        Ehsan Yaghoubi ; Diana Borza ; Aruna Kumar ; Hugo Proença
COMMENTS:       To be published in IEEE international conference on image processing 2020 (ICIP 2020). Codes are available at https://github.com/Ehsan-Yaghoubi/reid-strong-baseline
HIGHLIGHT:      This paper introduces one 'implicit' attentional mechanism for deep learning frameworks, that provides simultaneously: 1) masks-free; and 2) foreground-focused samples for the inference phase.

35, TITLE:      CrackGAN: Pavement Crack Detection Using Partially Accurate Ground Truths Based on Generative Adversarial Learning
http://arxiv.org/abs/1909.08216
AUTHORS:        Kaige Zhang ; Yingtao Zhang ; Heng-Da Cheng
HIGHLIGHT:      To tackle this problem, we propose crack-patch-only (CPO) supervised generative adversarial learning for end-to-end training, which forces the network to always produce crack-GT images while reserves both crack and BG-image translation abilities by feeding a larger-size crack image into an asymmetric U-shape generator to overcome the "All Black" issue.

36, TITLE:      The Explanation Game: Explaining Machine Learning Models Using Shapley Values
http://arxiv.org/abs/1909.08128
AUTHORS:        Luke Merrick ; Ankur Taly
HIGHLIGHT:      In this work, we illustrate how subtle differences in the underlying game formulations of existing methods can cause large differences in the attributions for a prediction.

37, TITLE:      Giving Commands to a Self-driving Car: A Multimodal Reasoner for Visual Grounding
http://arxiv.org/abs/2003.08717
AUTHORS:        Thierry Deruyttere ; Guillem Collell ; Marie-Francine Moens
COMMENTS:       Updated acknowledgements
HIGHLIGHT:      We propose a new spatial memory module and a spatial reasoner for the Visual Grounding (VG) task.

38, TITLE:      Survey on Evaluation Methods for Dialogue Systems
http://arxiv.org/abs/1905.04071
AUTHORS:        Jan Deriu ; Alvaro Rodrigo ; Arantxa Otegi ; Guillermo Echegoyen ; Sophie Rosset ; Eneko Agirre ; Mark Cieliebak
HIGHLIGHT:      In this paper we survey the methods and concepts developed for the evaluation of dialogue systems.

39, TITLE:      Variational Denoising Network: Toward Blind Noise Modeling and Removal
http://arxiv.org/abs/1908.11314
AUTHORS:        Zongsheng Yue ; Hongwei Yong ; Qian Zhao ; Lei Zhang ; Deyu Meng
COMMENTS:       11 pages, 4 figures
HIGHLIGHT:      In this work we propose a new variational inference method, which integrates both noise estimation and image denoising into a unique Bayesian framework, for blind image denoising.

40, TITLE:      A Two stage Adaptive Knowledge Transfer Evolutionary Multi-tasking Based on Population Distribution for Multi/Many-Objective Optimization
http://arxiv.org/abs/2001.00810
AUTHORS:        Zhengping Liang ; Weiqi Liang ; Xiuju Xu ; Ling Liu ; Zexuan Zhu
COMMENTS:       14 pages, 8 figures, 7 tables, 61 references
HIGHLIGHT:      To further investigate the effectiveness of EMT-PD on many-objective optimization problems, a multi-tasking many-objective test suite is also designed in this paper.

41, TITLE:      A survey on Semi-, Self- and Unsupervised Learning in Image Classification
http://arxiv.org/abs/2002.08721
AUTHORS:        Lars Schmarje ; Monty Santarossa ; Simon-Martin Schröder ; Reinhard Koch
HIGHLIGHT:      In this survey, we provide an overview of often used ideas and methods in image classification with fewer labels.

42, TITLE:      EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation
http://arxiv.org/abs/1912.11951
AUTHORS:        Roshan Dathathri ; Blagovesta Kostova ; Olli Saarikivi ; Wei Dai ; Kim Laine ; Madanlal Musuvathi
HIGHLIGHT:      This paper presents a new FHE language called Encrypted Vector Arithmetic (EVA), which includes an optimizing compiler that generates correct and secure FHE programs, while hiding all the complexities of the target FHE scheme.

43, TITLE:      A Universal Representation Transformer Layer for Few-Shot Image Classification
http://arxiv.org/abs/2006.11702
AUTHORS:        Lu Liu ; William Hamilton ; Guodong Long ; Jing Jiang ; Hugo Larochelle
HIGHLIGHT:      We analyze variants of URT and present a visualization of the attention score heatmaps that sheds light on how the model performs cross-domain generalization.

44, TITLE:      Leveraging Frequency Analysis for Deep Fake Image Recognition
http://arxiv.org/abs/2003.08685
AUTHORS:        Joel Frank ; Thorsten Eisenhofer ; Lea Schönherr ; Asja Fischer ; Dorothea Kolossa ; Thorsten Holz
COMMENTS:       Accepted to ICML 2020. New experiments, updated several sections, code: https://github.com/RUB-SysSec/GANDCTAnalysis
HIGHLIGHT:      In this paper, we address this shortcoming and our results reveal that in frequency space, GAN-generated images exhibit severe artifacts that can be easily identified.

45, TITLE:      Self-Supervised Joint Learning Framework of Depth Estimation via Implicit Cues
http://arxiv.org/abs/2006.09876
AUTHORS:        Jianrong Wang ; Ge Zhang ; Zhenyu Wu ; XueWei Li ; Li Liu
HIGHLIGHT:      In this work, we propose a novel self-supervised joint learning framework for depth estimation using consecutive frames from monocular and stereo videos.

46, TITLE:      Deep Networks as Logical Circuits: Generalization and Interpretation
http://arxiv.org/abs/2003.11619
AUTHORS:        Christopher Snyder ; Sriram Vishwanath
HIGHLIGHT:      We present a hierarchical decomposition of the DNN discrete classification map into logical (AND/OR) combinations of intermediate (True/False) classifiers of the input.

47, TITLE:      Algorithm for Computing Approximate Nash equilibrium in Continuous Games with Application to Continuous Blotto
http://arxiv.org/abs/2006.07443
AUTHORS:        Sam Ganzfried
HIGHLIGHT:      We present a new algorithm for computing Nash equilibrium strategies in continuous games.