Evolution of Heuristics for Give-Away Checkers

The efficacy of two evolutionary approaches to the problem of generation of heuristical linear and non-linear evaluation functions in the game of give-away checkers is tested in the paper. Experimental results show that both tested methods lead to heuristics of reasonable quality and evolutionary algorithms can be successfully applied to heuristic generation in case not enough expert knowledge is available. 1 The game of Give-Away Checkers (GAC) The game of US give-away checkers [1] is played according to the same rules as US checkers. The way of determining the winner is the only exception. In order to win in the game of GAC a player has to lose all his/her pieces or be unable to perform a move. The rules of the game are simple and widely known and - at the same time - the results of brute-force algorithm using trivial strategy of losing pieces as quickly as possible are unsatisfactory. One of the reasons for unsuitability of this simple algorithm is the fact that a single piece is ba...

• Type: Conference
• Year: 2005
• Where: ICANN
• Authors: Magdalena Kusiak, Karol Waledzik, Jacek Mandziuk
• Keywords: Game, Give-away Checkers, ICANN 2005, Non-linear Evaluation Functions