On The Evolution Of Strategies In The Population Of Iterated Prisoner's Dilemma Game Players

Since the beginning of human society,cooperation has always been a common but important behavior.Although individuals behave motivated by self-interest,there is a wide range of cooperative behavior in the society where individuals are gathered.This phenomenon of cooperation in the population of selfish individuals has attracted scholars in many fields to study the evolution of cooperation.Evolutionary game theory provides a powerful theoretical framework for this issue.Current researches on the evolution of cooperation mainly focus on the game model and the mechanism to promote cooperation.Scholars have proposed models such as Prisoner's Dilemma Game and Snowdrift Game that describe the contradiction between individual's interest and group's interest,and also proposed five rules to promote the evolution of cooperation such as direct reciprocity,indirect reciprocity and spatial reciprocity,etc.However,this research area is still in its infancy;and further explorations are demanded.Based on the existing problems,this thesis proposes a strategy based on the memory mechanism that adapts to the noisy environment,i.e.,Tit-For-Tat with Cumulative Tolerance.In this thesis,this strategy firstly proposed,based on the description of its underlying concepts.Then the noise adaptability and evolution of this strategy are tested through two kinds of repeated game experiments under noisy environment.One is between two TFTCTs and the other is between TFTCT and the existing classical strategies.The experimental results show that TFTCT performs better with the increasing of the noise level,and it is an advantageous strategy in the evolution who can invade other strategies.Afterwards,the research focuses on the evolutionary model of strategies population and carries out an evolutionary iterated prisoner's dilemma game on the well-mixed population.The results indicate that the evolutionary model of the population evolves from stochastic decisionmaking to defection,and ultimately reaches a state of stable cooperation.And TFTCT is the motivation for the emergence of cooperation in the evolution.Finally,the same game is performed again on the two-dimensional grid structure to observe whether the population structure will influence the population's evolutionary model.The simulation results show that under the action of TFTCT,the population on the grid presents the same evolutionary pattern as on the well-mixed population.The research in this thesis enriches the understanding of the formation and evolution mechanism of cooperation theoretically.For social practice,the understanding of cooperation is deepened through the study of the evolutionary model of cooperation in the population.And it also has reference value in inspiring organizations to develop cooperation in fierce competition.