STRATEGY OPTIMIZATION METHODS FOR TWO-PLAYER GAMES WITH GENETIC ALGORITHMS

D. . Trofimov, A. A. Shalyto


Read the full article 

Abstract

The article deals with genetic algorithm modification, allowing to solve the optimization problem without setting objective function evidently with the use of comparison function of two candidate solutions. This is achieved by applying the so-called tournament systems. Suggested algorithm enables to use the basic genetic programming principles in problems where it is impossible or too inefficient. For example, the task of an optimal strategy creation against a random opponent in two-player games refers to such problems class.


Keywords: genetic algorithms, two-player games
Copyright 2001-2017 ©
Scientific and Technical Journal
of Information Technologies, Mechanics and Optics.
All rights reserved.

Яндекс.Метрика