Co-evolutionary learning in the n-choice iterated prisoner's dilemma with PSO algorithm in a spatial environment

Abstract

The evolution of strategies in n-choice iterated prisoner's dilemma game is studied on spatial environment. This paper presents and investigates the application of co-evolutionary training techniques based on particle swarm optimization (PSO) to evolve cooperation, and exploring different parameter configurations via numerical simulations. Key model parameters include the size of the population, the interaction topology, the number of choices and the cost-to-benefit ratio. The simulation results reveal that the spatial structure does promote higher levels of cooperative behaviors, the cost-to-benefit ratio and the multiple choices are important factors in determining the strategy evolution.