Study On Mechansims Of Cooperation Evolution Of Network Groups Based On Individual Dynamic Attributes

Cooperation is a ubiquitous phenomenon in nature.It is an important principle to promote the evolution of biological population and human society.How the cooperative behavior emerges and evolves in a competitive group?It is an interdisciplinary issue that attracts the attention of scholars in various fields.In the social-technical network systems,users form the networked group for the same target.They cooperate to complete crowd sensing,distributed computing and realize resource sharing.However,in the autonomous and dynamic network environments,most rational individuals never contribute.Their selfish behavior results in the lack of resources,low efficiency and trust crisis of the collaboration systems.Therefore many researchers focus the emergence of cooperation among the netw ork group.In order to guarantee the quality of service of multi-agent systems,we need to further understand the rules of network group behavior and build effective incentive mechanisms to promote cooperation.Under the theoretical framework of spatial evolutionary game,this dissertation studies the evolution of network group cooperation behavior based on individual dynamic attributes.We focus on the favorable factors that affect the emergence and maintenance of cooperation and provide a basis for the design of incentive mechanisms of collaborative system.The research is mainly from the following three aspects:First,a multi-strategy Prisoner's Dilemma game model in multi-layer network has been constructed.We study the evolution of cooperative behavior based on individual reputation attribute in static network.A modified conservative criterion based on the second order social norm has been proposed.To some extend,it reduces the impact of reputation misjudgement on the evolution of cooperation.In addition,the research is foucus on three layer coupling structure:including trust network,interacting network and learning network.We study the evolution dynamics of group cooperative behavior on such networks.The theoretical analysis and simulation results show that the indirect reciprocity mechanism can further encourage cooperation with help of network reciprocity.The sturctures of trust and interacting networks determine the evolution of the strategy,while the learning network only affects the spead of strategy among network group.Then a heterogeneous resource investment mechanism based on age attribute has been proposed in the game of space public goods.We study the inherent mechanisms of natural attribute driven emergence of cooperationin in static networks.After defining lifetime attributes in different application scenarios,we design a static age protocol QueA and dynamic age protocols,such as NatA and StrA.The results show that when the age of the individual is positively related to the amount of investment,compared to the basic mechanism ignoring the age factors,the three rules can promote the emergence of cooperationto some extends.Especially,StrA rule helps form of a beneficial age structure for cooperation in the networked group.At last but the most important,co-evolution models based on the label of reputation and age respectively have been constructed.We study the partner selection mechanism based on individual dynamic attributes.And it drives the co-evolution of group behavior and the structure.Taking the misjudgement in reputation management into account,the structure evolution modes based on the global and local reputations have been designed.And they ensure the system maintains a high level of cooperation.In addition,we discuss the novel mode to update social relations considering age factors.In the direct weighted network,the relationship strength is measured by the income and age difference of two players.The simulation and analysis results show that the above mechanisms can effectively promote the emergence of cooperation in the network group by forming a stable cooperative cluster.To sum up,the research enriches the theory of cooperative evolution based on indirect reciprocity and space reciprocity.It helps to understand the internal mechanism of the emergence of altruistic behavior in the network group.Also the dissertation can provide the basis and useful suggestion for the design of the incentive mechanism of the cooperative system.