PSEUDO-OPTIMAL STRATEGIES IN NO-LIMIT POKER

Article type: Research Article

Authors: Andersson, Rickard¹

Affiliations: Umea, Sweden

Note: [1] Umea University, Department of Computer Science, Umea, Sweden. Email: [email protected].

Abstract: Games have always been a strong driving force in Artificial Intelligence. In the last ten years huge improvements has been made in perfect information games like chess and Othello. The strongest computer agents can nowadays beat the strongest human players. This is not the case for imperfect information games such as poker and bridge where creating an expert computer player has shown to be much harder. Previous research in poker has either addressed limit poker or simplified variations of poker games. This paper tries to extend known techniques successfully used in limit poker to no-limit. No-limit poker increases the size of the game tree drastically. To reduce the complexity an abstracted model of the game is created. Finding an optimal strategy for the new model is now a minimization problem using linear programming techniques. The result is a set of pseudo-optimal strategies for no-limit Texas Hold’em. A bot named AGGROBOT was built from these strategies which perform well as long as the players’ stack sizes are fairly small.

DOI: 10.3233/ICG-2006-29303

Journal: ICGA Journal, vol. 29, no. 3, pp. 143-149, 2006