cvpr 注意力机制_今日论文|DeepMind强化学习工具包模拟关节机制&微软Lumos自动诊断度量回归&NLP论文贡献度量&元学习最优性&MORE...

于 2020-12-19 03:01:00 发布
阅读量1k 收藏 1
点赞数 1
文章标签： cvpr 注意力机制
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_39861823/article/details/112536832
版权
这篇博客汇总了CVPR 2020会议上关于计算机视觉和自然语言处理的最新研究。涵盖深度学习在图像分割、目标检测、图像生成、神经网络理论及强化学习等多个领域的进展。文章介绍了如BEV-Seg、多尺度图卷积网络、LAMP等新方法，并探讨了NLP中如机器阅读、领域适应、词义诱导等议题。此外，还讨论了用于诊断和评估的工具库Lumos，以及无监督对话评估与强化学习的应用。

摘要生成于 C知道 ，由 DeepSeek-R1 满血版支持， 前往体验 >

6月24日人工智能领域新增论文 271篇，AI日读精选其中 48篇推荐给大家。这些论文主要来自NeurIPS，ICML，CVPR，ECML，WACV，BMVC，IJCAI，JCDL，AIST，ICCIT，SIGdial等会议与期刊。其中包含 计算机视觉论文14篇[1-14]， 自然语言处理论文4篇[15-18]， 方法论论文11篇[22-32]， 神经网络原理论文3篇[33-35]， 语音技术论文1篇[42]， 知识图谱论文1篇[43]， 强化学习论文3篇[44-46]， 推荐系统论文2篇[47-48]，点击文末

28609a50ea6fc91740ea13b13836bd65.png

计算机视觉

Computer Vision

[1]

BEV-Seg: Bird's Eye View Semantic Segmentation Using Geometry and Semantic Point Cloud

Mong H. Ng, Kaahan Radia, Jianfei Chen, Dequan Wang, Ionel Gog, Joseph E. Gonzalez
摘 要：

bd65874e12749d52bef0d772b9a79cf5.png

原 文：http://arxiv.org/pdf/2006.11436v2

[2]

A Multiscale Graph Convolutional Network Using Hierarchical Clustering

Alex Lipov, Pietro Liò
摘 要：

0016283635a03c771e9e8a997c70ea58.png

原 文：http://arxiv.org/pdf/2006.12542v1

[3]

LAMP: Large Deep Nets with Automated Model Parallelism for Image Segmentation

Wentao Zhu, Can Zhao, Wenqi Li, Holger Roth, Ziyue Xu, Daguang Xu
摘 要：

e6684699f10f21a0655ca42b21dfe227.png

原 文：http://arxiv.org/pdf/2006.12575v1

[4]

NeuralScale: Efficient Scaling of Neurons for Resource-Constrained Deep Neural Networks

Eugene Lee, Chen-Yi Lee
摘 要：

b7ea6183b5e047847dd73efa5849ec32.png

原 文：http://arxiv.org/pdf/2006.12813v1

资 源：github.com/eugenelet/NeuralScale

[5]

SLV: Spatial Likelihood Voting for Weakly Supervised Object Detection

Ze Chen, Zhihang Fu, Rongxin Jiang, Yaowu Chen, Xian-sheng Hua
摘 要：

dc252cd1b09348433710fc025e28d5ad.png

原 文：http://arxiv.org/pdf/2006.12884v1

[6]

C-SURE: Shrinkage Estimator and Prototype Classifier for Complex-Valued Deep Learning

Yifei Xing, Rudrasis Chakraborty, Minxuan Duan, Stella Yu
摘 要：

38cab20bea4fd3c2fe1180bf82d99c64.png

原 文：http://arxiv.org/pdf/2006.12590v1

[7]

Extension of Direct Feedback Alignment to Convolutional and Recurrent Neural Network for Bio-plausible Deep Learning

Donghyeon Han, Gwangtae Park, Junha Ryu, Hoi-jun Yoo
摘 要：

466be69e580b675b5ad2e06d47a18191.png

原 文：http://arxiv.org/pdf/2006.12830v1

[8]

PAC-Bayes Analysis Beyond the Usual Bounds

Omar Rivasplata, Ilja Kuzborskij, Csaba Szepesvari, John Shawe-Taylor
摘 要：

b36ac90e87a694a1c9d45c80b2a44e90.png

原 文：http://arxiv.org/pdf/2006.13057v1

[9]

Instant 3D Object Tracking with Applications in Augmented Reality

Adel Ahmadyan, Tingbo Hou, Jianing Wei, Liangkai Zhang, Artsiom Ablavatski, Matthias Grundmann
摘 要：

df3b691d670706a0b1d6fa7a065ab16f.png

原 文：http://arxiv.org/pdf/2006.13194v1

[10]

DCNNs: A Transfer Learning comparison of Full Weapon Family threat detection forDual-Energy X-Ray Baggage Imagery

A. Williamson, P. Dickinson, T. Lambrou, J. C. Murray
摘 要：

bc1602d0c4cd4e7ead5ee8b40d3a803e.png

原 文：http://arxiv.org/pdf/2006.13065v1

[11]

AFDet: Anchor Free One Stage 3D Object Detection

Runzhou Ge, Zhuangzhuang Ding, Yihan Hu, Yu Wang, Sijia Chen, Li Huang, Yuan Li
摘 要：

d721ada603ed8af07a254dba3e8e2090.png

原 文：http://arxiv.org/pdf/2006.12671v1

[12]

Distributed Subgraph Enumeration via Backtracking-based Framework

Zhaokang Wang, Weiwei Hu, Chunfeng Yuan, Rong Gu, Yihua Huang
摘 要：

03476545e2c1a9e217ba33538d880b63.png

原 文：http://arxiv.org/pdf/2006.12819v1

资 源：github.com/PasaLab/BENU

[13]

Deep Learning of Unified Region, Edge, and Contour Models for Automated Image Segmentation

Ali Hatamizadeh
摘 要：

f3633cfcff569f4c35821fe5e3b47c1d.png

原 文：http://arxiv.org/pdf/2006.12706v1

[14]

PoseGAN: A Pose-to-Image Translation Framework for Camera Localization

Kanglin Liu, Qing Li, Guoping Qiu
摘 要：

c8946fb9e46cec1f20e0ddeec290aca3.png

原 文：http://arxiv.org/pdf/2006.12712v1

自然语言处理

Natural Language Processing

[15]

NLPContributions: An Annotation Scheme for Machine Reading of Scholarly Contributions in Natural Language Processing Literature

Jennifer D'Souza, Sören Auer
摘 要：

5cc6c8d815e684579f9cb5989079163c.png

原 文：http://arxiv.org/pdf/2006.12870v1

资 源：github.com/jenlindadsouza/

NLPContributions

[16]

Domain Adaptation for Semantic Parsing

Zechang Li, Yuxuan Lai, Yansong Feng, Dongyan Zhao
摘 要：

5cba48b9ccbe5fdfab3fe2b5b57f570b.png

原 文：http://arxiv.org/pdf/2006.13071v1

[17]

Combining Neural Language Models for WordSense Induction

Nikolay Arefyev, Boris Sheludko, Tatiana Aleksashina
摘 要：

8f8afc1786714e2fe11acb6f2709a429.png

原 文：http://arxiv.org/pdf/2006.13200v1

[18]

Unsupervised Evaluation of Interactive Dialog with DialoGPT

Shikib Mehri, Maxine Eskenazi
摘 要：

ede6d3c24cface0cbe252904c1e2a52a.png

原 文：http://arxiv.org/pdf/2006.12719v1

资 源：github.com/shikib/fed

数据集

Dataset

[19]

MANTRA: A Machine Learning reference lightcurve dataset for astronomical transient event recognition

Mauricio Neira, Catalina Gómez, John F. Suárez-Pérez, Diego A. Gómez, Juan Pablo Reyes, Marcela Hernández Hoyos, Pablo Arbeláez, Jaime E. Forero-Romero
摘 要：

c0bf06e3889fafb28d96effaa28fc8d1.png

原 文：http://arxiv.org/pdf/2006.13163v1

[20]

RP2K: A Large-Scale Retail Product Dataset forFine-Grained Image Classification

Jingtian Peng, Chang Xiao, Xun Wei, Yifan Li
摘 要：

0fefdb4a7591b32cfc9173e8910a3a8e.png

原 文：http://arxiv.org/pdf/2006.12634v1

任务与挑战

Task & Challenge

[21]

CLC: Complex Linear Coding for the DNS 2020 Challenge

Hendrik Schröter, Tobias Rosenkranz, Alberto N. Escalante-B., Andreas Maier
摘 要：

6e087b3b190a00d5b63545d8170f254f.png

原 文：http://arxiv.org/pdf/2006.13077v1

方法论

Methodology

[22]

On the Global Optimality of Model-Agnostic Meta-Learning

Lingxiao Wang, Qi Cai, Zhuoran Yang, Zhaoran Wang
摘 要：

5036cbfceb01c3df83da64646f87e928.png

原 文：http://arxiv.org/pdf/2006.13182v1

[23]

Contrastive Generative Adversarial Networks

Minguk Kang, Jaesik Park
摘 要：

ed65eb1fd8875bf6600d80e24a49f3f3.png

原 文：http://arxiv.org/pdf/2006.12681v1

资 源：github.com/POSTECH-CVLab/PyTorch-StudioGAN

[24]

Logical Neural Networks

Ryan Riegel, Alexander Gray, Francois Luus, Naweed Khan, Ndivhuwo Makondo, Ismail Yunus Akhalwaya, Haifeng Qian, Ronald Fagin, Francisco Barahona, Udit Sharma, Shajith Ikbal, Hima Karanam, Sumit Neelam, Ankita Likhyani, Santosh Srivastava
摘 要：

b222ffa9b69063f8d9e2b477028b272a.png

原 文：http://arxiv.org/pdf/2006.13155v1

[25]

Inductive Unsupervised Domain Adaptation for Few-Shot Classification via Clustering

Xin Cong, Bowen Yu, Tingwen Liu, Shiyao Cui, Hengzhu Tang, Bin Wang
摘 要：

2ce321e44846c2d24664b836ebe201fa.png

原 文：http://arxiv.org/pdf/2006.12816v1

[26]

Can you tell? SSNet -- a Sagittal Stratum-inspired Neural Network Framework for Sentiment Analysis

Apostol Vassilev, Munawar Hasan
摘 要：

7901a006f5a7f6d2fa7fa77e54555037.png

原 文：http://arxiv.org/pdf/2006.12958v1

[27]

not-MIWAE: Deep Generative Modelling with Missing not at Random Data

Niels Bruun Ipsen, Pierre-Alexandre Mattei, Jes Frellsen
摘 要：

02c630f57ccc1b11b3142172a0701f28.png

原 文：http://arxiv.org/pdf/2006.12871v1

资 源：github.com/AudeSportisse/stat

[28]

Limits of Transfer Learning

Jake Williams, Abel Tadesse, Tyler Sam, Huey Sun, George D. Montanez
摘 要：

f4b5ca92dade047a055824cd8a1f0c1b.png

原 文：http://arxiv.org/pdf/2006.12694v1

[29]

Prediction error-driven memory consolidation for continual learning. On the case of adaptive greenhouse models

Guido Schillaci, Luis Miranda, Uwe Schmidt
摘 要：

c73be5194e7de25a7c92e913a9cb5942.png

原 文：http://arxiv.org/pdf/2006.12616v1

[30]

Parameter Estimation Bounds Based on the Theory of Spectral Lines

Arnab Sarker, Joseph E. Gaudio, Anuradha M. Annaswamy
摘 要：

37ebc0ade723cb5e568d2eb923fb3f1f.png

原 文：http://arxiv.org/pdf/2006.12687v1

[31]

Perceptual Adversarial Robustness: Defense Against Unseen Threat Models

Cassidy Laidlaw, Sahil Singla, Soheil Feizi
摘 要：

f58d9d357c58b20e527de52e76c79937.png

原 文：http://arxiv.org/pdf/2006.12655v1

[32]

Developing a Mathematical Negotiation Mechanism for a Distributed Procurement Problem and a Hybrid Algorithm for its Solution

Zohreh Kaheh, Reza Baradaran Kazemzadeh, Ellips Masehian, Ali Husseinzadeh Kashan
摘 要：

a5844e6c61cf179f372d258e155d0782.png

原 文：http://arxiv.org/pdf/2006.13140v1

神经网络原理

Neural Network Theory

[33]

Density-embedding layers: a general framework for adaptive receptive fields

Francesco Cicala, Luca Bortolussi
摘 要：

db4059421df3a33eef2b1cae923e1f91.png

原 文：http://arxiv.org/pdf/2006.12779v1

[34]

RayS: A Ray Searching Method for Hard-label Adversarial Attack

Jinghui Chen, Quanquan Gu
摘 要：

73816a8a820af13b970eba7835c7812c.png

原 文：http://arxiv.org/pdf/2006.12792v1

[35]

Simple and Effective VAE Training with Calibrated Decoders

Oleh Rybkin, Kostas Daniilidis, Sergey Levine
摘 要：

adabe0af4eb63b4717cc973c7d94a2da.png

原 文：http://arxiv.org/pdf/2006.13202v1

综述

Survey

[36]

Artificial Intelligence-Assisted Energy and Thermal Comfort Control for Sustainable Buildings: An Extended Representation of the Systematic Review

Ghezlane Halhoul Merabet, Mohamed Essaaidi, Mohamed Ben-Haddou, Basheer Qolomany, Junaid Qadir, Muhammad Anan, Ala Al-Fuqaha, Riduan Mohamed Abid, Driss Benhaddou
摘 要：

2f7dffa1a0f2d5d51d68db08bd09bc12.png

原 文：http://arxiv.org/pdf/2006.12559v1

[37]

A Survey on Deep Learning for Localization and Mapping: Towards the Age of Spatial Machine Intelligence

Changhao Chen, Bing Wang, Chris Xiaoxuan Lu, Niki Trigoni, Andrew Markham
摘 要：

c2e7b8d767146548ca9e14e40934245b.png

原 文：http://arxiv.org/pdf/2006.12567v1

[38]

A survey of repositories in graph theory

Srinibas Swain, C. Paul Bonnington, Graham Farr, Kerri Morgan
摘 要：

f4460f99a94f0a0cb5e449d9e4bb1fb7.png

原 文：http://arxiv.org/pdf/2006.12741v1

平台与工具

Platform & Tool

[39]

Lumos: A Library for Diagnosing Metric Regressions in Web-Scale Applications

Jamie Pool, Ebrahim Beyrami, Vishak Gopal, Ashkan Aazami, Jayant Gupchup, Jeff Rowland, Binlong Li, Pritesh Kanani, Ross Cutler, Johannes Gehrke
摘 要：

e894ca3112913f30d818113a3c0a1039.png

原 文：http://arxiv.org/pdf/2006.12793v1

资 源：github.com/andosa/treeinterpreter, github.com/linkedin/luminol, github.com/microsoft/MS-Lumos

[40]

dm_control: Software and Tasks for Continuous Control

Yuval Tassa, Saran Tunyasuvunakool, Alistair Muldal, Yotam Doron, Siqi Liu, Steven Bohez, Josh Merel, Tom Erez, Timothy Lillicrap, Nicolas Heess
摘 要：

10bafd865a3bd4ae518a4c17bbe263fc.png

原 文：http://arxiv.org/pdf/2006.12983v1

资 源：github.com/deepmind/dm_control

[41]

hxtorch: PyTorch for ANNs on BrainScaleS-2

Philipp Spilger, Eric Müller, Arne Emmel, Aron Leibfried, Christian Mauch, Christian Pehle, Johannes Weis, Oliver Breitwieser, Sebastian Billaudelle, Sebastian Schmitt, Timo C. Wunderlich, Yannik Stradmann, Johannes Schemmel
摘 要：

59ace7c9d6176666f5d4904724dffd2f.png

原 文：http://arxiv.org/pdf/2006.13138v1

语音技术

Audio & Speech

[42]

Unsupervised Sound Separation Using Mixtures of Mixtures

Scott Wisdom, Efthymios Tzinis, Hakan Erdogan, Ron J. Weiss, Kevin Wilson, John R. Hershey
摘 要：

684d15116b112752a14b6d650bd2e0ae.png

原 文：http://arxiv.org/pdf/2006.12701v1

知识图谱

Knowledge Graph

[43]

Iterative Deep Graph Learning for Graph Neural Networks: Better and Robust Node Embeddings

Yu Chen, Lingfei Wu, Mohammed J. Zaki
摘 要：

4025aadb662ecea5630b590602597d73.png

原 文：http://arxiv.org/pdf/2006.13009v1

强化学习

Reinforcement Learning

[44]

ELSIM: End-to-end learning of reusable skills through intrinsic motivation

Arthur Aubret, Laetitia Matignon, Salima Hassas
摘 要：

02911ea2e620a406499dc2ce02f6b46c.png

原 文：http://arxiv.org/pdf/2006.12903v1

资 源：github.com/maximecb/gym-minigrid

[45]

Combinatorial Pure Exploration of Dueling Bandit

Wei Chen, Yihan Du, Longbo Huang, Haoyu Zhao
摘 要：

e9af03d9656b7c6011168cb7d0187f0b.png

原 文：http://arxiv.org/pdf/2006.12772v1

[46]

Online Multi-agent Reinforcement Learning for Decentralized Inverter-based Volt-VAR Control

Haotian Liu, Wenchuan Wu
摘 要：

55e550597ad8f709f44e1b1c88c1ce68.png

原 文：http://arxiv.org/pdf/2006.12841v1

推荐系统

Recommendation System

[47]

Hybrid Session-based News Recommendation using Recurrent Neural Networks

Gabriel de Souza P. Moreira, Dietmar Jannach, Adilson Marques da Cunha
摘 要：

dceb35ca704bd0c8de581b257f6d9aeb.png

原 文：http://arxiv.org/pdf/2006.13063v1

[48]

Better User Recommendations using Enhancing Software Development Process Repository

Ziaur Rahman, Md. Kamrul Hasan
摘 要：

f18523607dc9115dfed7c1a11a9f7e75.png

原 文：http://arxiv.org/pdf/2006.12738v1

81c8f3c775e2c5772290aaf9ac5f84c0.png

点个在看，再下论文~
(密码：6tvc)

008f11fbf1cadae8b9b389f08d347c85.png
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值