Genetic Algorithm for Static Games with N players

Abstract

One of the most important problems in the Game Theory is the possibility to obtain multiple equilibrium points when the concavity condition is not fulfilled. This means that algorithms that use gradient techniques have convergence problems or in the worst case cannot be used to find all the equilibrium points. The complexity increases with the number of players and the number of actions for every player. In this paper an alternative solution to find the equilibrium points of a static game with constraints using Genetic Algorithm is proposed. A convergence analysis of the algorithm is presented with some examples for the case of N-players.