Research On The Evolution Of Cooperation And Incentive Mechanisms On Complex Networks From A Heterogeneous Perspective

Cooperative behavior is the basis of the development of human society,and it exists widely in human society.However,from a perspective of biological evolution,why cooperative behavior can exist widely is still an unsolved mystery.Therefore,people explore incentive mechanisms constantly which can promote cooperation in order to manage the world better.It is one of powerful methods to study cooperative behavior of evolutionary game on complex networks for people.From the perspective of heterogeneity and based on evolutionary game on complex networks,this dissertation designs some incentive mechanisms that can promote cooperation for prisoner's dilemma game,snowdrift game and public goods game,respectively.In addition,to explain the evolution of cooperation,the influence of heterogeneous factors on strategy selection is studied.Specific studies are as follows:Firstly,this dissertation studies the impact of heterogeneous updating mechanisms on the evolution of cooperation among enterprises in the context of collaborative innovation in emerging strategic industry.The relationship between enterprises is modeled by the prisoner's dilemma game and the snowdrift game.Moreover,we design three heterogeneous updating mechanisms according to inertia,strategy preference and teaching ability.Enterprises in the network are divided into two categories randomly:the classical Fermi updating rule is used for the first category,and the heterogeneous updating mechanism is for the second one.We analyze the influence of the above three factors on strategy selection of enterprises.Heterogeneous inertia updating mechanism considers the phenomenon that enterprises tend to stick to the strategy they used in the past.The simulation results show that the fraction of enterprises adopting cooperation increases with the increasing of the proportion of using heterogeneous inertia updating mechanism when enterprises' inertia is relatively small.There is an optimal proportion to maximize the proportion of cooperation when inertia is large.Heterogeneous strategy preference updating mechanism means that enterprises is influenced by strategy preference when they change strategy.Results show that this mechanism is effective for the larger temptation of defection when enterprises do not distinguish between cooperation and defection.On the contrary,the fraction of enterprises adopting cooperation increases obviously when enterprises distinguish strategies.Teaching ability refers to the ability that an enterprise influences its neighbors.This factor is considered by a coevolutionary model.The results show that the willingness of enterprises choosing cooperation turns stronger influenced by teaching ability,and the ability of enterprises to resist the temptation of defection is improved Furthennore,the above three heterogeneous updating mechanisms are tested on heterogeneous networks and the results show that they can also promote cooperation.Secondly,this dissertation studies the heterogeneous investment mechanism and tolerance-based punishment mechanism in spatial public goods game.The amount of individual investment will become different completely when individuals use heterogeneous investment mechanism.Results show that heterogeneous investment mechanism can enhance network reciprocity if investment intensity is positive,and it can promote cooperation on different networks.At the same time,the average investment of cooperators under the heterogeneous investment mechanism is lower than that of the traditional spatial public goods game.On the contrary,the average investment of cooperators increases and the evolution of cooperation is restrained.It gives the opportunity of punishment to cooperators when individuals use tolerance-based punishment mechanism.Results show that it not only can reduce the frequencies and costs of using punishment,but also increase the proportion of cooperators if individuals use tolerance-based punishment mechanism.In addition,the effect of punishment mechanism based on same tolerance is more effective than that of heterogeneous tolerance.Finally,combining data sharing problem in P2P media network,this dissertation studies the impact of multiple punishment strategies on the evolution of users' behavior based on the above studies.The model assumes that the cooperator can decide whether to punish or not according to the tolerance of the group,and there are many punishment strategies to use.However,cooperators have to bear the cost equally irrespective of their strategies if punishment is decided to implement.This seting can overcome the second-order free riding.Results show that the punishment mechanism can promote the evolution of cooperation under specific conditions.As the number of free riders(defectors)decreases,the competition among uploaders(cooperators)becomes more intense.In most cases,users tend to use a single punishment strategy in stable state,but there are also multiple punishment strategies coexisting.Moreover,under the same conditions,with the increase of the number of punishment strategies,the proportion of cooperators is also increasing.