Evolution of Self-Interested Agents: An Experimental Study

Naoki Yamada and Chiaki Sakama

Proceedings of the 7th Multi-Disciplinary International Workshop on Artificial Intelligence (MIWAI), Lecture Notes in Artificial Intelligence 8271, Springer-Verlag, pages 329-340, 2013.


In this paper, we perform an experimental study to examine the evolution of self-interested agents in cooperative agent societies. To this end, we realize a multiagent system in which agents initially behave altruistically by sharing information of food. After generations of a genetic algorithm, we observe the emergence of selfish agents who do not share food information. The experimental results show the process of evolving self-interested agents in resource-restrictive environments, which is observed in nature and in human society.

Full Paper (PDF 443K) Slide