Dynamic Model Making for Iterated Prisoner's Dilemma Based on Markov Chain

Consider the bounded rationality of players in evolutionary games and the character of Markov chain, this paper presented a dynamic model for iterated prisoner's dilemma game. In this model, strategies choosing of players was mapped to a Markov decision process with two states, and this model can successfully simulate strategies choosing and learning of players. It can be used not only in finite populations but also in infinite populations.