Title:Automata representation of successful strategies for social dilemmas

Abstract:A social dilemma appears when defection is individually favored than cooperation although the latter is collectively optimal. A class of successful strategies of direct reciprocity were recently found for the iterated prisoner's dilemma [Yi et al., J. Theor. Biol. (2017)] and for the three-person iterated public goods game [Murase and Baek, J. Theor. Biol. (2018)]. A successful strategy forms a cooperative Nash equilibrium in the presence of implementation error while assuring that the long-term payoff never becomes less than the co-players' regardless of their strategies, when the error rate is small. Although we have a list of actions prescribed by the successful strategies, the rationale behind them have not been fully understood. In this paper, we wish to provide a better understanding for the successful strategies by converting them into equivalent automata with a minimal number of states. Each of the states becomes interpretable in this representation, whereby one can gain insight into the underlying mechanism.