Learning a Game Strategy Using Pattern-Weights and Self-play

Abstract. This paper demonstrates the use of pattern-weights in order to develop a strategy for an automated player of a non-cooperative version of the game of Diplomacy. Diplomacy is a multi-player, zerosum and simultaneous move game with imperfect information. Patternweights represent stored knowledge of various aspects of a game that are learned through experience. An automated computer player is developed without any initial strategy and is able to learn important strategic aspects of the game through self-play by storing pattern-weights and using temporal difference learning.

• Type: Journal
• Year: 2002
• Where: CG
• Authors: Ari Shapiro, Gil Fuchs, Robert Levinson
• Keywords: CG 2002, Computer Graphics, Game, Important Strategic Aspects, Non-cooperative Version