• May 15, 2019 News!Vol.7, No.5- Vol.8, No.4 has been indexed by EI (Inspec).   [Click]
  • Aug 01, 2018 News! [CFP] 2020 the annual meeting of IJMO Editorial Board, ECDMO 2020, will be held in Athens, Greece, February 15-17, 2020.   [Click]
  • Sep 30, 2019 News!Vol 9, No 6 has been published with online version. 12 original aritcles from 6 countries are published in this issue.    [Click]
General Information
    • ISSN: 2010-3697  (Online)
    • Abbreviated Title: Int. J. Model. Optim.
    • Frequency: Bimonthly
    • DOI: 10.7763/IJMO
    • Editor-in-Chief: Prof. Adrian Olaru
    • Executive Editor: Ms.Yoyo Y. Zhou
    • Abstracting/ Indexing: ProQuest, Crossref, Electronic Journals Library, Google Scholar, EI (INSPEC, IET), EBSCO, etc.
    • E-mail ijmo@iacsitp.com
Editor-in-chief
Prof. Adrian Olaru
University Politehnica of Bucharest, Romania
I'm happy to take on the position of editor in chief of IJMO. It's a journal that shows promise of becoming a recognized journal in the area of modelling and optimization. I'll work together with the editors to help it progress.
IJMO 2019 Vol.9(3): 140-145 ISSN: 2010-3697
DOI: 10.7763/IJMO.2019.V9.699

Opponent Modelling with Eligibility Trace for Multi-agent Reinforcement Learning

Hao Chen, Jian Huang, and Jianxing Gong
Abstract—Markov games and reinforcement learning algorithms are applied successfully in multi-agent learning systems such as Minimax-Q. Because of the interdependence between agents, it’s time consuming to find the optimal policy when agents learning concurrently. Some algorithms accelerate convergences through spatial or action generalization, which requires domain-dependent prior knowledge. In order to improve learning efficiency directly, the opponent modelling Q(λ) algorithm is proposed which combines fictitious play in game theory and eligibility trace in reinforcement learning. A series of empirical evaluations were conducted in the classical soccer domain. Compared with several other algorithms, it is proved that the algorithm contributed in this paper significantly enhances the learning performance of multi-agent systems.

Index Terms—Opponent modelling, markov Games, multi-agent, reinforcement learning.

The authors are with the College of Artificial Intelligence, National University of Defense Technology, Hunan, Changsha 410073 China. (e-mail: nudtchenhao15a@163.com, nudtjHuang@hotmail.com, fj_gjx@qq.com).

[PDF]

Cite: Hao Chen, Jian Huang, and Jianxing Gong, "Opponent Modelling with Eligibility Trace for Multi-agent Reinforcement Learning," International Journal of Modeling and Optimization vol. 9, no. 3, pp. 140-145, 2019.

Copyright © 2008-2019. International Journal of Modeling and Optimization. All rights reserved.
E-mail: ijmo@iacsitp.com