Intelligent Poker Player
Orr Bernstein (ob29@cornell.edu), Jonathan Margulies (jm392@cornell.edu), Cliff Tsai (ct247@cornell.edu)
Cornell University, Department of Computer Science

Introduction

Problem Definition and Algorithm

Program Design

Experimental Evaluation

Conclusion

Future Work

References
Introduction

We created a program that plays limit hold'em poker, the most common form of poker in American casinos. Our goal is to evaluate different poker-playing models and to hopefully find which algorithms and heuristics are most effective.

A poker shell was created that allows different strategies to be snapped in against each other. We designed many different strategies that employ different methods, advantages, and disadvantages and experimented by playing them against each other under different circumstances. The results we found are typical of poker: Though there are some general rules regarding the best strategy, in a game that involves so much adjustment and trickery there can be no single "best" strategy. The key is knowing what strategy to use at any given time.