ICLR2020106篇深度强化学习顶会论⽂汇总
深度强化学习实验室报道
转载⾃： EndtoEnd.ai
编辑：DeepRL硼元素符号
【导读】今年的ICLR⼤会转到了线上举⾏，DeepMind和哈佛的研究⼈员投稿了⼀篇神经⽹络控制虚拟⼩⽩⿏模的论⽂⼗分亮眼。此次ICLR⼤会，华⼈学者参与论⽂数占⽐近60%，Google⼊选80余篇表现依旧抢眼，⽽国内的研究团队也不落下风，满分论⽂频现。本届ICLR 2020共有2594篇投稿，687 篇被接收。其中：48篇 oral 108篇，spotlights 531篇， poster 录取率为 26.5%，相⽐去年的31.4% 略有降低。强化学习⼀直是ICLR投稿的热点，近年来强化学习及深度强化学习不断刷新着⼈类在游戏、棋牌等领域的最好成绩，关于⾕歌研究⼈员⽤6⼩时完成AI芯⽚设计，也是采⽤了深度强化学习⽅法，强化学习的威⼒不容⼩觑。本⽂共列举了106篇深度强化学习领域的论⽂。
排名1
平均得分8
论⽂地址
标题 Dynamics-aware Unsupervid Skill Discovery
得分 8 8 8
Variance0
Decision Accept (Talk)
排名1
平均得分8
论⽂地址
标题Contrastive Learning Of Structured World Models
得分 8 8 8
Variance0
Decision Accept (Talk)
排名1
平均得分8
论⽂地址
标题Implementation Matters In Deep Rl: A Ca Study On Ppo And Trpo
得分 8 8 8
Variance0
Decision Accept (Talk)
排名1
平均得分8
论⽂地址
红军过雪山标题Gendice: Generalized Offline Estimation Of Stationary Values
得分 8 8 8
Variance0
Decision Accept (Talk)
排名1
平均得分8
论⽂地址
标题Causal Discovery With Reinforcement Learning
得分 8 8 8
Variance0
Decision Accept (Talk)
阴茎勃起长度
排名2
平均得分7.33
论⽂地址
标题Is A Good Reprentation Sufficient For Sample Efficient Reinforcement Learning?得分 8 8 6
Variance0.89
Decision Accept (Spotlight)
排名2
你以为你是谁
平均得分7.33
论⽂地址
标题Harnessing Structures For Value-bad Planning And Reinforcement Learning
得分 6 8 8
Variance0.89
Decision Accept (Talk)
排名2
平均得分7.33
论⽂地址
标题Explain Your Move: Understanding Agent Actions Using Focud Feature Saliency
得分 6 8 8
Variance0.89
Decision Accept (Poster)
排名2
平均得分7.33
论⽂地址
标题Meta-q-learning
得分 8 8 6
Variance0.89
Decision Accept (Talk)
排名2
平均得分7.33
论⽂地址
标题Discriminative Particle Filter Reinforcement Learning For Complex Partial Obrvations 得分 8 6 8
Variance0.89
Decision Accept (Poster)
排名2
平均得分7.33
论⽂地址
标题Disagreement-regularized Imitation Learning
哥特式建筑风格得分 6 8 8
Variance0.89
Decision Accept (Spotlight)
排名2
平均得分7.33
论⽂地址
标题Doubly Robust Bias Reduction In Infinite Horizon Off-policy Estimation
得分 6 8 8
Variance0.89
Decision Accept (Spotlight)
排名2
平均得分7.33
论⽂地址
标题Seed Rl: Scalable And Efficient Deep-rl With Accelerated Central Inference 得分 8 6 8
优秀简历模板下载Variance0.89
Decision Accept (Talk)
排名2
平均得分7.33
论⽂地址
标题The Ingredients Of Real World Robotic Reinforcement Learning
得分 6 8 8
Variance0.89
Decision Accept (Spotlight)
夸奖女人的词语排名2
平均得分7.33
论⽂地址
标题Watch The Unobrved: A Simple Approach To Parallelizing Monte Carlo Tree Search 得分 8 6 8
Variance0.89
Decision Accept (Talk)
排名2
平均得分7.33
论⽂地址
标题Meta-learning Acquisition Functions For Transfer Learning In Bayesian Optimization 得分 8 6 8
Variance0.89
Decision Accept (Spotlight)
排名2
平均得分7.33
论⽂地址
中文符号标题 A Clor Look At Deep Policy Gradients
得分 8 6 8
Variance0.89
Decision Accept (Talk)
排名2
平均得分7.33
论⽂地址
标题Fast Task Inference With Variational Intrinsic Successor Features
得分 8 6 8
Variance0.89
Decision Accept (Talk)
