Simulating the Evolution of Behavior: the Iterated Prisoners’ Dilemma Problem
David M. Chess’
Comp uting Systems Department,
IBM T. J. Watson Research Cent er, Post Office Box 218, Yorktown Heights, NY, USA
A bs t r a ct.
A system is described in which a number of art ificial agents (represented by simple mathematical expressions) compe te for t he right to «re produce» (that is, to cause new agents with similar properties to be genera ted). By simulat ing some of the essential features of biological evolution, the system makes possible some novel insights into the behavior of communi ties of agents over time. The results of Fujiki and Dickinson on the Iterated Prisoners’ Dilemma problem (IPD) are essentially confirmed. The ty pical course of evolution of a community of IPD players is descri bed) and possibilities for further work are outlined . This study is also relevant to machine learning) an d adapti ve systems in general.