Matrix games with nonuniform payoff distributions. (English) Zbl 0993.91001

Summary: The theoretical analysis of the statistical properties of 2-person zero-sum games with random payoff matrices is generalized to payoff matrices with elements whose average and variance depend on the column they belong to. The value of the game and the distribution of the strategies are solved analytically using methods from statistical mechanics of neural networks. The analytical results are confirmed by simulations.


91A05 2-person games
82C32 Neural nets applied to problems in time-dependent statistical mechanics
Full Text: DOI


[1] Kanter, I.; Saad, D., Phys. Rev. Lett., 83, 2660 (1999)
[2] Kanter, I.; Kanter, E.; Ein-Dor, L., Europhys. Lett., 51, 244 (2000)
[3] Berg, J.; Engel, A., Phys. Rev. Lett., 81, 1999 (1998)
[4] Challet, D.; Zhang, Y. C., Physica, A 246, 407 (1997)
[5] Ein-Dor, L.; Metzler, R.; Kanter, I.; Kinzel, W., Phys. Rev., E 63, 066103 (2001)
[6] von Neumann, J.; Morgenstern, O., Theory of Games and Economic Behavior (1953), Princeton University Press: Princeton University Press Princeton · Zbl 0053.09303
[7] Jones, A. J., Game Theory: Mathematical Models of Conflict (1980), Ellis Horwood Limited · Zbl 0419.90086
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.