Genetic Algorithm for Iterated Prisoner’s Dilemma
In iterated prisoner's dilemma, two players repeatedly play Prisoner’s dilemma. In this game, opponent can be punished by non-cooperating for his defect in previous rounds. We need to implement a strategy that can eventually reduce the payoffs received over multiple runs. A player strategy predict the next move of the opponent. A player can choose to cooperate, defect or move based on his opponent’s move or even tit for tat/2tats.