Real-time imitation-based adaptation of gaming behaviour in modern computer games

In the course of the recent complexification and sophistication of commercial computer games, the creation of competitive artificial players that are able to behave intelligently and successfully in the featured highly dynamic and complex virtual worlds has become a considerable challenge. This paper describes an evolutionary realtime adaptation approach to produce competitive artificial players in an action game. The proposed method is inspired by the idea of social learning or cultural evolution. Thus, the agents try to adapt to the level of their opponents by the exchange of information about advantageous behaviours within the population. In addition, the behaviour of the opponents and other players is recorded and used to create more sophisticated and human-like agents. Categories and subject descriptors: I.2.11 [Artificial Intelligence]: Distributed Artificial Intelligence— Intelligent agents, Multiagent systems; I.2.6 [Artificial Intelligence] Learning—Knowledge Acquis...

• Type: Conference
• Year: 2008
• Where: GECCO
• Authors: Steffen Priesterjahn, Alexander Weimer, Markus Eberling
• Keywords: Artificial Intelligence, Competitive Artificial Players, Evolutionary Realtime Adaptation, GECCO 2008, Optimization