Exponential moving average Q-learning algorithm
Authors:
- 1 April 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 16 (23251824) , 31-38
Abstract
A multi-agent policy iteration learning algorithm is proposed in this work. The Exponential Moving Average (EMA) mechanism is used to update the policy for a Q-learning agent so that it converges to an optimal policy against the policies of the other agents. The proposed EMA Q-learning algorithm is examined on a variety of matrix and stochastic games. Simulation results show that the proposed algorithm converges in a wider variety of situations than state-of-the-art multi-agent reinforcement learning (MARL) algorithms.Keywords
This publication has 9 references indexed in Scilit:
- Effective learning in the presence of adaptive counterpartsJournal of Algorithms, 2009
- A Multiagent Reinforcement Learning Algorithm with Non-linear DynamicsJournal of Artificial Intelligence Research, 2008
- A Comprehensive Survey of Multiagent Reinforcement LearningIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2008
- AWESOME: A general multiagent learning algorithm that converges in self-play and learns a best response against stationary opponentsMachine Learning, 2006
- Multi-Agent Reinforcement Learning: A Survey2006
- Multiagent learning using a variable learning rateArtificial Intelligence, 2002
- 10.1162/jmlr.2003.4.6.1039Applied Physics Letters, 2000
- Evolution and learning in multiagent systemsInternational Journal of Human-Computer Studies, 1998
- Reinforcement Learning: A SurveyJournal of Artificial Intelligence Research, 1996