Research On The Evolutionary Game Mechanism Of Complex Networks Based On Dynamic Structure And Dynamic Payoff

Many complex systems exist in the real world,such as transportation networks,social networks,financial markets,etc.Complex networks are highly abstracted from complex systems,which represent individuals within real complex systems and the associations between individuals in terms of nodes and connected edges,while evolutionary game theory provides a powerful theoretical paradigm for describing the behavior and interactions of individuals.With the development of social economy,cooperation has become one of the necessary conditions for people to successfully achieve their goals and improve their quality of life.Cooperation can bring additional benefits and can help build good relationships among group members.The study of evolutionary game mechanisms on complex networks is important for understanding the causes of cooperative behavior in social games.Most of the traditional evolutionary game studies focus on using complex network modeling and adopting certain game models to simulate real situations and solve practical problems,but the research on the mechanisms that promote cooperation and the inner laws of group evolution is more limited.At the same time,most previous studies have been conducted on static network models,while the real network structure is often dynamic and uncertain,and the relationship between individuals may change at any time,which can affect the behavior and decision making of individuals and thus the group cooperation rate.In order to further explore the mechanism of promoting cooperation and enrich the research,this paper conducts computer simulation experiments based on NW small world network,under the weak prisoner’s dilemma game model and public goods game model with heterogeneous investment,and discusses in detail the evolutionary game mechanism of complex networks based on dynamic structure and dynamic payoff.Specific work and innovations are as follows.(1)This paper proposes a dynamic payoff reward and punishment model with a temporal cycle,which includes three parameters of reward factor,punishment factor and payoff influence period .When a node’s strategy shifts from betrayal to cooperation,we call it has a tendency to cooperation and reward its payoff for evolutionary time steps,similarly when a node’s strategy shifts from cooperation to betrayal,we call it a cooperation distancer and penalize its payoff for evolutionary time steps.The results suggest that,in contrast to the traditional reward-free complex network evolution game,either rewarding those who converge on cooperation alone or punishing those who diverge from cooperation alone can contribute to the emergence of cooperation.The promotion effect of the reward and punishment factor on cooperation is amplified by the introduction of the payoff influence period .The network cooperation rate under both models is positively correlated with .Then we further explored the changes in the hot zone of group cooperation rate on the parameter space with the reward and punishment factors as the horizontal and vertical coordinates,and found that the influence of the reward factor on group cooperation rate increase is greater than that of the punishment factor when takes a small value,while both will possess a high influence on group cooperation rate increase when increases to a certain value.Meanwhile,we learn from the analysis of the network evolution process that in the prisoner’s dilemma game,the effect of the change of on the steady-state cooperation rate is approximately linear,while in the public goods game there is a mutation threshold of ,which can cause the cooperation rate to change abruptly from low to high.In addition,the increase of can shorten the time for the cooperation rate to enter the steady state,and the system will reach the convergence state more quickly.(2)To simulate the social tendencies of individuals in real networks,this paper constructs a dynamic network topology model with cooperator preferences,where all nodes update their neighbor relationships with probability after completing each round of the game,and the node disconnects from a betrayed neighbor while adding a new connection with a random neighbor of the cooperating neighbor.The experimental results show that the dynamic network topology mechanism can effectively improve the group cooperation rate under different game models,and there is an optimal action interval for .The value of within the interval can maximize the cooperation rate,and when exceeds the threshold value continues to increase it will have a certain inhibitory effect on the cooperation.In summary,this paper conducted several simulation experiments by Monte Carlo simulation method to verify that both dynamic payoff reward and punishment mechanism and dynamic network topology mechanism can significantly enhance the group cooperation rate in the evolutionary game on complex networks,providing some theoretical guidance for promoting the emergence of cooperative behavior in problems of business competition,social governance,environmental protection and other social fields.