Game Mechanism Research Under Co-evolution Of Complex Network And Evolutionary Dynamics

Game theory is a subject focusing on behaviors and decisions of rational players involvedin competitive and interactive activities. We can found principles of game theory behindalmost every interactive behavior in the nature world as well as human society. In the game,players aimed at maximum their earnings or profits. Though it will pull down player’s profitsand lead them into disadvantage, cooperative behaviors are still extensively exist in real world.Then why cooperation thrive in a cruel world of natural selection has become a puzzle everhaunting the researcher. Evolutionary game theory and especially, the introduction of thecomplex networks provided a new perspective for understanding the cooperation behavior.Evolutionary games on the complex networks have become one of the central topical ofgame theory. Game model, strategy update rule and network model three factors consistsevolutionary games on network. Networks and game theory are incorporated like this: nodesin the network represent players of a particular game; edges represents its two endpoints areneighbors. Most evolutionary games on the complex networks evolve as follow describe: first,players play game with their neighbors and accumulate payoff; then, using specific updaterule, update their strategy with reference of their payoff; at last, repeat the progress until thestrategy of player became steady.This paper introduced development process and present situation of game theory and thebasic knowledge of complex networks. Then, studied update rules and network topology,gained innovative products described as follow:1) Apply Moran rule, unconditional imitation rule and replicator role on snow drift game. Tostudy how the change of update rule affects evolutionary games. And then, abandoningusual idea of sole update rule in the network, we introduced an evolution process usingtwo update rules. The outcomes of our model pointed out that, update rule have asignificant impact on the cooperation level.2) Studied co-evolution behavior of public goods game. Introduced evolutionary preferentialattachment (EPA) model, which acted as a cannel between game evolution and networktopology. In the EPA model, the network structure has been formed by game evolutionand in turn, promoted cooperation level in the game. Studies on EPA shows that3) Constructed a new network training model named evolutionary elimination andreproduction. Bad players will be isolated and then eliminated and substituted by a newnode. This model provided a co-evolution mechanism so that outcome of game will shapethe network; gradually transform it from a homogenous network to a heterogeneous oneand increase universal cooperation level.