Application of Hidden Markov Models in Game Theory

• Eduardo R. Waghabi UFRJ
• Mario R. F. Benevides UFRJ

Abstract

In this paper we discuss possibilities of connections between two different areas of research: Game Theory and Hidden Markov Models. Two different models are presented: the first one shows some traps one must avoid, while presenting positive results on the use of Genetic Algorithms for the HMM training. The second model shows success on the use of HMM for strategic advantages on mixed strategy games. This model is generalized in the last section aiming its use on a wider variety of mixed strategy games.

References

CHAU, C. W. , KWONG, S. , DIU, C.K. ET AL., 1997, “Optimization Of HMM By A Genetic Algorithm”, Conferência Internacional IEEE sobre Acústica, Fala e Processamento de Sinal, Munique, Alemanha.

FUDENBERG, D., TIROLE, J., 1994, “Game Theory”, MIT Press, Cambridge.

GINTIS, H., 2000, “Game Theory Evolving”, Princeton University Press, New Jersey, EUA.

JURAFSKY, D., MARTIN, J. H., 2006, “Speech and Language Processing: An introduction to natural language processing”, Prentice Hall, New Jersey, EUA, 1st ed.

NASH, J. F., 1950, “Non-Cooperative Games”, Dissertação de Doutorado, Universidade de Princeton, New Jersey, EUA.

RABINER, L., 1989, “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition”, Proceedings of the IEEE, 77 (2), p. 257–286.

RUSSELL, S., NORVIG, P., 2004, “Inteligência Artificial”, Elsevier Editora, São Paulo, Brasil, tradução da Segunda Edição.

VON NEUMANN, J., MORGENSTERN, O., 1944, “Theory of Games and Economic Behavior”, Princeton University Press, New Jersey, EUA, 60.ed.