Research In Public Goods Game Model Based On Reputation In Complex Networks

Individuals in complex system can cooperate,distribute,adapt,communicate,organize and replicate,the cooperation driving force between individuals can be modelled according to the Evolution Game Theory.Evolution Game Theory,as a key means to research complex system,is not only convenient to study the cooperation between agents qualitatively and quantitatively,but also easy for establishing graceful models.Therefore,in the field of complexity science,evolution game dynamics caused the attention of the whole complex system researchers.Firstly,in our society,interaction often occurs in similar individuals because of individuals' ability limit in itself,and individuals are easily affected by the environment,therefore,modelling for population structures and studying the impact on evolution is an important question.Secondly,the phenomenon that individuals help others will be more likely to be helped is widespread,how to establish effective reputation mechanism based on the indirect reciprocity in complex systems and promote the evolution of cooperation is another important question.Based on the above two existing questions,this paper combines the complex networks and evolutionary game theory to study the dynamics process of individual behavior in complex networks evolution,so that puts forward two kinds of reputation mechanism to improve the cooperation level.The main research content is as follows:Firstly,in well-mixed networks,each individual has chance to interact with any one individual,defection will be the Nash Equilibrium strategy because defector can share payoff without giving some investment.But in our society,individuals always interact with similar individuals and occasionally interact with individuals who are very different.So this paper introduces reputation in selection process,individuals are divided into different interaction groups according to their reputation score,individuals interact with individuals in the same group each generation.At the same time,former reputation mechanism only increase or decrease according to the individuals' action,and this paper proposes the reputation update combining individual behavior and group behavior,each group is graded into different classes of group behavior and adopts different rules to update individuals' reputation,tolerance-based update rules are proposed to identify unusual individuals.Secondly,there are diverse population structure in real society,individuals always interact with some fixed neighbors,so this paper simulate this phenomenon and study the evolution of cooperation in different complex networks.Defectors still have an advantage in spatial evolution game,because defector can share payoff without giving some investment.So this paper extends the existing reputation mechanism and let reputation be the judgment condition of individuals' participating in the public goods game.An individual can refuse to participate in a public goods game if the group reputation is so far below individual's reputation.Consequently cooperators avoid defectors' invasion according to this reputation-based judgment,and the cooperation are greatly evolved.Tolerance-based reputation mechanism is proved to be effective by a great many of experimentations.