Evolutionary algorithm against player for the game

Abstract

The computer games have been offered as a new Drosophila for AI research. Many games are a test bed for the developments of ideas in computer game playing programs. This paper describes how we use Genetic Algorithm to control the level of difficulty in a game based on a player's skill. The game learns how to offer appropriate difficulty for players in our researches. We use Factor of Difficulty Control in a game(FDC) as Chromosomes and Factor of Users' Adaptation to a game(FUA) as Fitness in the Genetic Algorithm. We propose to control the difficulty of a game based on the players' skill fundamentally using Genetic Algorithm. Then we apply it to the puzzle game, and prove that it decreases the gap between beginners and experts. Thus we try to improve previous Genetic Algorithm to optimize the game. In this paper, we propose an Evolutionary Algorithm against Player for game (E.A.P) that controls the difficulty of a game based on the player's propensity and proficiency by Genetic Algorithm (G.A). We suggest a game algorithm that enables a game to change the difficulties by itself based on the player's suitability to the game using Genetic algorithm.