Opponent's style modeling based on situations for bayesian poker

Authors:
Ruobing Li;Wenkai Li;Lin Shang;Yang Gao;Mengjie Zhang
Affiliations:
State Key Laboratory for Novel Software Technology, Nanjing University, China;State Key Laboratory for Novel Software Technology, Nanjing University, China;State Key Laboratory for Novel Software Technology, Nanjing University, China;State Key Laboratory for Novel Software Technology, Nanjing University, China;School of Engineering and Computer Science, Victoria University of Wellington, New Zealand
Venue:
AI'12 Proceedings of the 25th Australasian joint conference on Advances in Artificial Intelligence
Year:
2012

Citing 6
Cited 0

Opponent modeling in poker

AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Using probabilistic knowledge and simulation to play poker

AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
The challenge of poker

Artificial Intelligence - Chips challenging champions: games, computers and Artificial Intelligence
Bayesian Artificial Intelligence

Bayesian Artificial Intelligence
Bayesian poker

UAI'99 Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence
Player modeling: Towards a common taxonomy

CGAMES '11 Proceedings of the 2011 16th International Conference on Computer Games

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a real poker game, one player can take actions of different styles in different situations. In this paper, a novel method is proposed to quantify and model the opponent's style in corresponding situation of a hand. Based on the proposed representation of Action Pair, the value of the style can be calculated and stored as "experience". When making a decision, the specific style will be obtained from the "experience". The style and the observable information will be used to estimate the value of the opponent's hand. In experiments, the obtained "experience" validates the correctness of our assumption that a player does not show an invariable style in all situations. The experimental results show that the agent player using our method can predict the value of the opponent's hand and earn more money in fixed hands comparing with the original agent.