QL2, a simple reinforcement learning scheme for two-player zero-sum Markov games: Proceedings of the 16th European Symposium on Artificial Neural Networks

Benoît Frénay, Marco Saerens

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférenceArticle dans les actes d'une conférence/un colloque

Résumé

Markov games is a framework which can be used to formalise n-agent reinforcement learning (RL). Littman (Markov games as a framework for multi-agent reinforcement learning, in: Proceedings of the 11th International Conference on Machine Learning (ICML-94), 1994.) uses this framework to model two-agent zero-sum problems and, within this context, proposes the minimax-Q algorithm. This paper reviews RL algorithms for two-player zero-sum Markov games and introduces a new, simple, fast. algorithm, called 2L(2).2L(2) is compared to several standard algorithms (Q-learning, Minimax and minimax-Q) implemented with the)ash library written in Python. The experiments show that 222 converges empirically to optimal mixed policies, as minimax-Q, but uses a surprisingly simple and cheap updating rule. (C) 2009 Elsevier B.V. All rights reserved.
langue originaleAnglais
titreProceedings of the 16th European Symposium on Artificial Neural Networks
Pages137-142
Nombre de pages6
Etat de la publicationPublié - 2009
Modification externeOui
Evénement16th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning, ESANN 2008 - Bruges, Belgique
Durée: 23 avr. 200825 avr. 2008

Une conférence

Une conférence16th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning, ESANN 2008
PaysBelgique
La villeBruges
période23/04/0825/04/08

Empreinte digitale Examiner les sujets de recherche de « QL2, a simple reinforcement learning scheme for two-player zero-sum Markov games: Proceedings of the 16th European Symposium on Artificial Neural Networks ». Ensemble, ils forment une empreinte digitale unique.

  • Contient cette citation

    Frénay, B., & Saerens, M. (2009). QL2, a simple reinforcement learning scheme for two-player zero-sum Markov games: Proceedings of the 16th European Symposium on Artificial Neural Networks. Dans Proceedings of the 16th European Symposium on Artificial Neural Networks (p. 137-142)