Deep Residual Reinforcement Learning (Extended Abstract)

被引：0

作者：

Zhang, Shangtong ^{[1
]}

Boehmer, Wendelin ^{[1
]}

Whiteson, Shimon ^{[1
]}

机构：

[1] Univ Oxford, Oxford, England

来源：

PROCEEDINGS OF THE THIRTIETH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2021 | 2021年

基金：

英国工程与自然科学研究理事会; 欧洲研究理事会;

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We revisit residual algorithms in both model-free and model-based reinforcement learning settings. We propose the bidirectional target network technique to stabilize residual algorithms, yielding a residual version of DDPG that significantly outperforms vanilla DDPG in commonly used benchmarks. Moreover, we find the residual algorithm an effective approach to the distribution mismatch problem in model-based planning. Compared with the existing TD(k) method, our residual-based method makes weaker assumptions about the model and yields a greater performance boost.

引用

页码：4869 / 4873

页数：5

共 50 条

[1] Robust Deep Reinforcement Learning with Adversarial Attacks Extended Abstract
Pattanaik, Anay
Tang, Zhenyi
Liu, Shuijing
Bommannan, Gautham
Chowdhary, Girish
[J]. PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS (AAMAS' 18), 2018, : 2040 - 2042
[2] Introspective Reinforcement Learning and Learning from Demonstration Extended Abstract
Li, Mao
Brys, Tim
Kudenko, Daniel
[J]. PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS (AAMAS' 18), 2018, : 1992 - 1994
[3] Deep Residual Attention Reinforcement Learning
Zhu, Hanhua
Kaneko, Tomoyuki
[J]. 2019 INTERNATIONAL CONFERENCE ON TECHNOLOGIES AND APPLICATIONS OF ARTIFICIAL INTELLIGENCE (TAAI), 2019,
[4] Deep Reinforcement Learning of Abstract Reasoning from Demonstrations
Clark-Turner, Madison
Begum, Momotaz
[J]. HRI '18: PROCEEDINGS OF THE 2018 ACM/IEEE INTERNATIONAL CONFERENCE ON HUMAN-ROBOT INTERACTION, 2018, : 160 - 168
[5] Deep Reinforcement Learning of Abstract Reasoning from Demonstrations
Clark-Turner, Madison
Begum, Momotaz
[J]. COMPANION OF THE 2018 ACM/IEEE INTERNATIONAL CONFERENCE ON HUMAN-ROBOT INTERACTION (HRI'18), 2018, : 372 - 372
[6] Stable Reinforcement Learning with Unbounded State Space (Extended Abstract)
Shah, Devavrat
Xie, Qiaomin
Xu, Zhi
[J]. LEARNING FOR DYNAMICS AND CONTROL, VOL 120, 2020, 120 : 581 - 581
[7] Interactive Relational Reinforcement Learning of Concept Semantics (Extended Abstract)
Nickles, Matthias
Rettinger, Achim
[J]. PROCEEDINGS OF THE TWENTY-FIFTH INTERNATIONAL CONFERENCE ON AUTOMATED PLANNING AND SCHEDULING, 2015, : 361 - 362
[8] Models for Autonomously Motivated Exploration in Reinforcement Learning (Extended Abstract)
Auer, Peter
Lim, Shiau Hong
Watkins, Chris
[J]. ALGORITHMIC LEARNING THEORY, 2011, 6925 : 14 - +
[9] Approximation of Convex Envelope Using Reinforcement Learning (Extended Abstract)
Borkar, Vivek S.
Akarsh, Adit
[J]. 2023 NINTH INDIAN CONTROL CONFERENCE, ICC, 2023, : 174 - 176
[10] Guiding Reinforcement Learning Exploration Using Natural Language Extended Abstract
Harrison, Brent
Ehsan, Upol
Riedl, Mark O.
[J]. PROCEEDINGS OF THE 17TH INTERNATIONAL CONFERENCE ON AUTONOMOUS AGENTS AND MULTIAGENT SYSTEMS (AAMAS' 18), 2018, : 1956 - 1958

← 1 2 3 4 5 →