Adaptive
Automata-based Model for Iterated n-Player’s Prisoner’s Dilemma
Sally Almanasra1, Khaled Suwais2 and Muhammad Rafie1
1School of Computer Sciences, Universiti Sains Malaysia, Malaysia
2Faculty of Computer Studies, Arab Open University, Saudi Arabia
Abstract: In this paper, we present a new technique of representing the player’s strategies by adaptive automata, which can handle complex strategies in large populations effectively. The representation the player’s strategies have a great impact on changing the player’s behaviour in rational environments. This model is built on the basis of changing the behaviour of the player’s gradually toward the cooperation. The gradualism is achieved by constructing three different adaptive automata at three different levels. The results showed that our model could represent the player’s strategies efficiently. The results proofed that the model is able to enhance the cooperation level between the participated player’s through few tournaments.
Keywords: Adaptive automata, prisoner’s dilemma, cooperative
behavior, INPPD.
Received
October 3, 2013; accepted June 9, 2014