A distributional perspective on value function factorization methods for multi-agent reinforcement learning

Wei Fang Sun; Cheng Kuang Lee; Chun Yi Lee

A distributional perspective on value function factorization methods for multi-agent reinforcement learning

Wei Fang Sun
, Cheng Kuang Lee
, Chun Yi Lee

Department of Electrical Engineering

National Tsing Hua University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Distributional reinforcement learning (RL) provides beneficial impacts for the single-agent domain. However, distributional RL methods are not directly compatible with value function factorization methods for multi-agent reinforcement learning. This work provides a distributional perspective on value function factorization, offering a solution for bridging the gap between distributional RL and value function factorization methods.

Original language	English
Title of host publication	20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021
Publisher	International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)
Pages	1659-1661
Number of pages	3
ISBN (Electronic)	9781713832621
State	Published - 2021
Event	20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021 - Virtual, Online Duration: 03 05 2021 → 07 05 2021

Publication series

Name	Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
Volume	3
ISSN (Print)	1548-8403
ISSN (Electronic)	1558-2914

Conference

Conference	20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021
City	Virtual, Online
Period	03/05/21 → 07/05/21

Bibliographical note

Publisher Copyright:
© 2021 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.

Keywords

Distributional RL
Multi-Agent RL
Reinforcement Learning

Cite this

Sun, W. F., Lee, C. K., & Lee, C. Y. (2021). A distributional perspective on value function factorization methods for multi-agent reinforcement learning. In 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021 (pp. 1659-1661). (Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS; Vol. 3). International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS).

Sun, Wei Fang ; Lee, Cheng Kuang ; Lee, Chun Yi. / A distributional perspective on value function factorization methods for multi-agent reinforcement learning. 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021. International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS), 2021. pp. 1659-1661 (Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS).

@inproceedings{f0ce455ac4954ebea393b8b5101845d8,

title = "A distributional perspective on value function factorization methods for multi-agent reinforcement learning",

abstract = "Distributional reinforcement learning (RL) provides beneficial impacts for the single-agent domain. However, distributional RL methods are not directly compatible with value function factorization methods for multi-agent reinforcement learning. This work provides a distributional perspective on value function factorization, offering a solution for bridging the gap between distributional RL and value function factorization methods.",

keywords = "Distributional RL, Multi-Agent RL, Reinforcement Learning",

author = "Sun, \{Wei Fang\} and Lee, \{Cheng Kuang\} and Lee, \{Chun Yi\}",

note = "Publisher Copyright: {\textcopyright} 2021 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.; 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021 ; Conference date: 03-05-2021 Through 07-05-2021",

year = "2021",

language = "英语",

series = "Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS",

publisher = "International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)",

pages = "1659--1661",

booktitle = "20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021",

address = "美国",

}

Sun, WF, Lee, CK & Lee, CY 2021, A distributional perspective on value function factorization methods for multi-agent reinforcement learning. in 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021. Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS, vol. 3, International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS), pp. 1659-1661, 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021, Virtual, Online, 03/05/21.

A distributional perspective on value function factorization methods for multi-agent reinforcement learning. / Sun, Wei Fang; Lee, Cheng Kuang; Lee, Chun Yi.
20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021. International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS), 2021. p. 1659-1661 (Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS; Vol. 3).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A distributional perspective on value function factorization methods for multi-agent reinforcement learning

AU - Sun, Wei Fang

AU - Lee, Cheng Kuang

AU - Lee, Chun Yi

PY - 2021

Y1 - 2021

N2 - Distributional reinforcement learning (RL) provides beneficial impacts for the single-agent domain. However, distributional RL methods are not directly compatible with value function factorization methods for multi-agent reinforcement learning. This work provides a distributional perspective on value function factorization, offering a solution for bridging the gap between distributional RL and value function factorization methods.

AB - Distributional reinforcement learning (RL) provides beneficial impacts for the single-agent domain. However, distributional RL methods are not directly compatible with value function factorization methods for multi-agent reinforcement learning. This work provides a distributional perspective on value function factorization, offering a solution for bridging the gap between distributional RL and value function factorization methods.

KW - Distributional RL

KW - Multi-Agent RL

KW - Reinforcement Learning

UR - https://www.scopus.com/pages/publications/85112264098

M3 - 会议稿件

AN - SCOPUS:85112264098

T3 - Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS

SP - 1659

EP - 1661

BT - 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021

PB - International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)

T2 - 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021

Y2 - 3 May 2021 through 7 May 2021

ER -

Sun WF, Lee CK, Lee CY. A distributional perspective on value function factorization methods for multi-agent reinforcement learning. In 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021. International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS). 2021. p. 1659-1661. (Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS).

A distributional perspective on value function factorization methods for multi-agent reinforcement learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Other files and links

Fingerprint

Cite this