Critic learning in multi agent credit assignment problem

被引：5

作者：

Rahaie, Zahra ^{[1
]}

Beigy, Hamid ^{[1
]}

机构：

[1] Sharif Univ Technol, Dept Comp Engn, Tehran, Iran

来源：

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS | 2016年 / 30卷 / 06期

关键词：

Multi-agent systems; credit assignment; reinforcement learning; interaction; history; knowledge; MODEL;

D O I：

10.3233/IFS-162093

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Multi-agent systems can be seen as an apparatus for testing the performance of real distributed systems. One problem encountered in multi-agent systems with the learning capability is credit assignment. This paper presents two methods for solving this problem. The first method assigns credit to the agents according to the history of the interaction while the second method assigns credit to the agents according to the knowledge of agents, and thus the shares of the agents are extracted from the feedback of the environment. The computer experiments show that critic learning has a positive impact in credit assignment problem.

引用

页码：3465 / 3480

页数：16

共 50 条

[21] Deep reinforcement learning with credit assignment for combinatorial optimization [J].

Yan, Dong ;

Weng, Jiayi ;

Huang, Shiyu ;

Li, Chongxuan ;

Zhou, Yichi ;

Su, Hang ;

Zhu, Jun .

PATTERN RECOGNITION, 2022, 124

[22] Organizational Learning as Credit Assignment: A Model and Two Experiments [J].

Fang, Christina .

ORGANIZATION SCIENCE, 2012, 23 (06) :1717-1732

[23] An Object Oriented Approach to Fuzzy Actor-Critic Learning for Multi-Agent Differential Games [J].

Schwartz, Howard .

2019 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (IEEE SSCI 2019), 2019, :183-190

[24] A multi-agent reinforcement learning using Actor-Critic methods [J].

Li, Chun-Gui ;

Wang, Meng ;

Yuan, Qing-Neng .

PROCEEDINGS OF 2008 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2008, :878-882

[25] Learning controlled and targeted communication with the centralized critic for the multi-agent system [J].

Sun, Qingshuang ;

Yao, Yuan ;

Yi, Peng ;

Hu, YuJiao ;

Yang, Zhao ;

Yang, Gang ;

Zhou, Xingshe .

APPLIED INTELLIGENCE, 2023, 53 (12) :14819-14837

[26] Learning controlled and targeted communication with the centralized critic for the multi-agent system [J].

Qingshuang Sun ;

Yuan Yao ;

Peng Yi ;

YuJiao Hu ;

Zhao Yang ;

Gang Yang ;

Xingshe Zhou .

Applied Intelligence, 2023, 53 :14819-14837

[27] Multi-agent off-policy actor-critic algorithm for distributed multi-task reinforcement learning [J].

Stankovic, Milos S. ;

Beko, Marko ;

Ilic, Nemanja ;

Stankovic, Srdjan S. .

EUROPEAN JOURNAL OF CONTROL, 2023, 74

[28] Solving the credit assignment problem: explicit and implicit learning of action sequences with probabilistic outcomes [J].

Wai-Tat Fu ;

John R. Anderson .

Psychological Research, 2008, 72 :321-330

[29] A Hierarchical Multi-Task and Multi-Agent Assignment Approach: Learning DQN Strategy From Execution [J].

Wang, Yu ;

Li, Huiping ;

Shen, Qingliang .

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, 2025, 22 :14712-14722

[30] Multi-Agent Credit Assignment and Bankruptcy Game for Improving Resource Allocation in Smart Cities [J].

Yarahmadi, Hossein ;

Shiri, Mohammad Ebrahim ;

Challenger, Moharram ;

Navidi, Hamidreza ;

Sharifi, Arash .

SENSORS, 2023, 23 (04)

← 1 2 3 4 5 →