网络演化博弈中结合收益与舆论的策略选择研究

Cooperation is widely existent in nature and human society. How does cooperation spontaneously emerge among selfish individuals? The evolutionary game theory provides a powerful theoretical framework to answer this question. In previous studies of the evolutionary game, individuals update their strategies according to payoff, that is, the goal of strategic adjustment is to maximize the payoff. However, in real life, the selection of strategy is often affected by public opinion. For example, people usually have conformist mentality, they tend to follow the majority opinion or behavior among their neighbors. In this project, we will combine payoff with public opinion to update individuals' strategies. Using different network structures (such as small-world networks and scale-free networks) and different game models (such as the prisoner’s dilemma game, the snowdrift game and the public goods game), we will study how the strategy updating based on the integration of payoff and public opinion affects the cooperation level. We expect our studies can deeply and systematically display the role of public opinion in networked evolutionary game. We want to understand the emergence of cooperation from a new insight. At the same time, we expect to reveal the internal relation between the evolutionary game and the opinion dynamics through our studies.

合作行为广泛存在于自然界和人类社会中。自私个体间的合作行为是如何自发涌现的？演化博弈为回答这一问题提供了强大的理论框架。在以往的演化博弈研究中，个体是基于收益进行策略更新的，即个体以自身收益最大化为目标来进行策略上的调整。然而在现实生活中，个体进行策略更新的时候，常常会受到社会舆论的影响。例如，人们常常有从众心理，倾向于追随周围邻居中占多数的意见或者行为。在本项目中，将结合收益与舆论，来决定个体的策略选择。我们将在不同的网络结构（如小世界网络和无标度网络等）及不同的博弈模型（如囚徒困境博弈、铲雪博弈和公共品博弈等）上，研究结合收益与舆论的策略选择对于系统合作水平的影响。我们期望通过本项目的研究，系统深入地呈现舆论在网络演化博弈中所起的作用，从一个全新的视角来理解合作行为的形成。同时，我们也期望通过本项目的研究，揭示演化博弈和舆论动力学之间的内在关联。

个体之间的相互协作是群体生存和发展的基础。到目前为止，演化博弈为理解合作行为的涌现提供了强大的理论框架。在本项目的研究中，个体基于收益与舆论，来决定是否选择合作策略。本项目的主要研究内容包括：基于舆论和收益的惩罚规则，基于舆论和收益的博弈参照对象选择，基于演化博弈的舆论同步研究。本项目综合运用复杂网络、统计物理、群体智能等理论方法，对不同网络结构（如方格网络、小世界网络和无标度网络）和不同博弈模型（如囚徒困境博弈、捐献博弈和公共品博弈）中的合作行为进行了系统深入的研究, 揭示了演化博弈和舆论动力学之间的内在关联。发表标注项目基金号的论文11篇（均被SCI收录），其中1篇论文入选ESI高被引论文，累计被SCI引用超过100次。联合培养博士2名，在读硕士研究生4名。