ADAPTIVE BEHAVIOUR FOR PRISONER DILEMMA STRATEGIES BASED ON AUTOMATA WITH MULTIPLICITIES

Cyrille Bertelle (1) , Marianne Flouret (2) , Véronique Jay (1) , Damien Olivier (1) , Jean-Luc Ponty (1)

1 RI2C - LITIS - Equipe Réseaux d'interactions et Intelligence Collective
2 LITIS - Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes

This paper presents the use of automata with multiplicities for the modelization of agent behaviours. Genetic algorithms can be defined on their probabilistic forms and are able to simulate adaptive behaviours. An application to a simulation of evolutive strategies in game theory is presented, specifically for the prisoner dilemma which is considered as a basic model of dynamic agents interactions in terms of cooperation and competition.

Mots clés: agent; automaton; behaviour; organization; multiplicities; genetic algorithm

Domaines: Intelligence artificielle [cs.AI]

Cyrille Bertelle, Marianne Flouret, Véronique Jay, Damien Olivier, Jean-Luc Ponty. ADAPTIVE BEHAVIOUR FOR PRISONER DILEMMA STRATEGIES BASED ON AUTOMATA WITH MULTIPLICITIES. ESS 2002, 14th European Simulation Symposium and Exhibition, Oct 2002, Dresden, Germany. ⟨hal-03317643⟩