Interactive Opponents Generate Interesting Games

In this paper we present experiments on neuroevolution mechanisms applied to predator/prey multi-character computer games. Our test-bed is a computer game where the prey (i.e. player) has to avoid its predators by escaping through an exit without getting killed. By viewing the game from the predators’ (i.e. opponents’) perspective, we attempt off-line to evolve neural-controlled opponents, whose communication is based on partial implicit information, capable of playing effectively against computer-guided fixed strategy players. However, emergent near-optimal behaviors make the game less interesting to play. We therefore discuss the criteria that make a game interesting and, furthermore, we introduce a generic measure of this category of (i.e. predator/prey) computer games’ interest (i.e. player’s satisfaction from the game). Given this measure, we present an evolutionary mechanism for opponents that keep learning from a player while playing against it (i.e. on-line) and we demonstrate its efficiency and robustness in increasing and maintaining the game’s interest. Computer game opponents following this on-line learning approach show high adaptability to changing player strategies which provides evidence for the approach’s effectiveness against human players.