Pattern Classification in No-Limit Poker: A Head-Start Evolutionary Approach
We have constructed a poker classification system which makes informed betting decisions based upon three defining features extracted while playing poker: hand value, risk, and aggressiveness. The system is implemented as a player-agent, therefore the goals of the classifier are not only to correctly determine whether each hand should be folded, called, or raised, but to win as many chips as possible from the other players. The decision space is found by evolutionary methods, starting from a data-driven initial state. Our results showed that evolving an agent from a data-driven “head-start” position resulted in the best performance over agents evolved from scratch, data-driven agents, random agents, and “always fold” agents.
KeywordsEvolution Pattern Classification No-limit Hold’em Poker
Unable to display preview. Download preview PDF.
- 2.Billings, D., Burch, N., Davidson, A., Holte, R., Schaeffer, J., Schauenberg, T., Szafron, D.: Approximating game theoretic optimal strategies for full-scale poker. In: International Joint Conference on Artificial Intelligence, August 2003, pp. 661–668 (2003)Google Scholar
- 3.Blank, T., Soh, L.K., Scott, S.: Creating an svm to play strong poker. In: International Conference on Machine Learning and Applications, December 2004, pp. 150–155 (2004)Google Scholar
- 4.Oliehoek, F.A., Vlassis, N., de Jong, E.D.: Coevolutionary nash in poker games. In: 17th Belgian-Dutch Conference on Artificial Intelligence, October 2005, pp. 188–193 (2005)Google Scholar
- 5.Southey, F., Bowling, M., Larson, B., Piccione, C., Burch, N., Billings, D., Rayner, C.: Bayes’ bluff: Opponent modelling in poker. In: Twenty-First Conference on Uncertainty in Artificial Intelligence, July 2005, pp. 550–558 (2005)Google Scholar
- 6.Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. Wiley Interscience, Hoboken (2000)Google Scholar