Abstraction Methods for Game Theoretic Poker
Abstraction is a method often applied to keep the combinatorial explosion under control and to solve problems of large complexity. Our work focuses on applying abstraction to solve large stochastic imperfect-information games, specifically variants of poker.We examine several different medium-size poker variants and give encouraging results for abstraction-based methods on these games.
Keywordspoker game theory imperfect information games Texas Hold’em
Unable to display preview. Download preview PDF.
- 1.Darse Billings, Denis Papp, Jonathan Schaeffer, and Duane Szafron. Opponent modeling in poker. In Proceedings of the 15th National Conference on Artificial Intelligence, pages 493–499, 1998.Google Scholar
- 2.Leslie Pack Kaelbling, Michael L. Littman, and Andrew W. Moore. Reinforcement learning: A survey. Journal of Artificial Intelligence Research, 4:237–285, 1996.Google Scholar
- 5.Avi Pfeffer, Daphne Koller, and Ken T. Takusagawa. State-space approximations for extensive form games. Workshop paper at FirstWorld Congress on Game Theory, 2000.Google Scholar
- 6.Richard S. Sutton and Andrew G. Barto. Reinforcement Learning: An Introduction. The MIT Press, 1998.Google Scholar
- 7.J. von Neumann and O. Morgenstern. Theory of Games and Economic Behavior. Princeton University Press, Princeton, NJ, 1947.Google Scholar