A multi-agent system integrating reinforcement learning, bidding and genetic algorithms

Article type: Research Article

Authors: Qi, Dehu | Sun, Ron

Affiliations: Lamar University, Computer Science Department, PO Box 10056, Beaumont, Texas, USA. E-mail: [email protected] | Rensselaer Polytechnic Institute, Cognitive Science Department, 110 Eighth Street, Carnegie 302A, Troy, New York 12180, USA. E-mail: [email protected]

Abstract: This paper presents a multi-agent reinforcement learning bidding approach (MARLBS) for performing multi-agent reinforcement learning. MARLBS integrates reinforcement learning, bidding and genetic algorithms. The general idea of our multi-agent systems is as follows: There are a number of individual agents in a team, each agent of the team has two modules: Q module and CQ module. Each agent can select actions to be performed at each step, which are done by the Q module. While the CQ module determines at each step whether the agent should continue or relinquish control. Once an agent relinquishes its control, a new agent is selected by bidding algorithms. We applied GA-based MARLBS to the Backgammon game. The experimental results show MARLBS can achieve a superior level of performance in game-playing, outperforming PubEval, while the system uses zero built-in knowledge.

Journal: Web Intelligence and Agent Systems: An international journal, vol. 1, no. 3-4, pp. 187-202, 2003