耦合网络的结构分析和演化动力学研究

发布时间：2018-11-17 21:33

【摘要】：在自然界和人类社会发展的过程中,博弈是一种普遍存在的现象。对博弈过程中博弈个体之间合作行为产生原因的解释成为了学术界的一个巨大挑战,而演化博弈理论为解决这一难题提供了一个全新的理论角度。近年来,网络科学理论研究发现复杂异构拓扑结构也是促进合作行为的一个重要机制。目前,大多数复杂网络上的博弈研究都集中在单层网络结构上,但现实世界复杂网络之间往往并不是相互孤立,而是相互依赖的。因此,研究耦合网络的结构特征及其演化动力学行为具有重要的现实意义。论文针对现实复杂系统耦合的关联特性,运用系统建模与仿真的方法,探索了多种耦合机制下双层耦合网络上囚徒困境博弈合作行为的演化。论文主要研究内容和创新性研究成果包括以下四个方面:1)研究了科研协作网络的社区结构。基于复杂网络方法,分析了科研协作网络的特征以及网络结构,进一步研究了科研协作网络的社区结构分布。结果表明科研协作网络具有明显的社区结构,为促进科研者合作提供了有意义的理论指导。2)提出了一个双层耦合网络模型,并探讨了在该网络模型上个体多样性对囚徒困境博弈的影响。模型中个体多样性具体表现为网络中博弈个体的自身策略传播能力不同;通过大量的计算机模拟仿真,研究结果表明,网络间的耦合以及个体多样性都能有效的促进网络中的合作行为,进而提高群体合作率。3)探讨了在加权耦合网络上囚徒困境博弈的演化行为。在加权耦合网络中,加权网络中博弈个体具有不同的权值,且服从三种随机分布,即均匀分布、幂律分布和指数分布,同时博弈个体权值的不同也体现了个体多样性。研究结果发现,网络之间的耦合可以大幅度提高群体的合作率,并且加权网络中合作率提升幅度明显大于标准无权网络。4)探索了耦合强度非均匀性对囚徒困境博弈行为的影响。论文提出了两种不同的非均匀耦合机制:对称耦合和不对称耦合;不同网络上相互依赖的两个个体的耦合强度相同,称为对称耦合;反之,称为不对称耦合。计算机模拟仿真结果表明,网络间的不均匀耦合会显著提高网络的整体合作水平,而且对称耦合机制比不对称耦合机制更能促进博弈个体间合作行为的传播。
[Abstract]:In the process of the development of nature and human society, game is a universal phenomenon. The explanation of the reasons for the cooperative behavior between individuals in the game process has become a great challenge to the academic community, and the evolutionary game theory provides a new theoretical angle to solve the problem. In recent years, the theory of network science has found that complex heterogeneous topology is also an important mechanism to promote cooperative behavior. At present, most of the game studies on complex networks focus on the single-layer network structure, but complex networks in the real world are often not isolated, but interdependent. Therefore, it is of great practical significance to study the structural characteristics and evolutionary dynamics of coupled networks. Aiming at the coupling characteristics of real complex systems, this paper uses the method of system modeling and simulation to explore the evolution of the cooperative behavior of prisoners' Dilemma game in multi-coupling networks. The main research contents and innovative research results are as follows: 1) the community structure of scientific research collaboration network is studied. Based on the complex network method, the characteristics and network structure of scientific research collaboration network are analyzed, and the community structure distribution of scientific research cooperation network is further studied. The results show that the research collaboration network has an obvious community structure, which provides a meaningful theoretical guidance for promoting scientific research cooperation. 2) A two-layer coupled network model is proposed. The influence of individual diversity on prisoner's dilemma game is discussed in this network model. The diversity of individuals in the model is embodied in the different communication ability of game players in the network. Through a large number of computer simulation results, the results show that the coupling between networks and individual diversity can effectively promote the cooperative behavior in the network. Then the cooperative rate of group is improved. 3) the evolutionary behavior of prisoners' dilemma game in weighted coupled networks is discussed. In the weighted coupled network, the game individual has different weights in the weighted network, and there are three random distributions, namely, uniform distribution, power law distribution and exponential distribution. At the same time, the individual diversity is also reflected by the difference of individual weights in the weighted network. The results show that the coupling between the networks can greatly improve the cooperation rate of the group. And the increase of cooperation rate in weighted networks is obviously greater than that in standard unweighted networks. 4) the influence of coupling intensity inhomogeneity on prisoners' dilemma game behavior is explored. In this paper, two different non-uniform coupling mechanisms are proposed: symmetric coupling and asymmetric coupling; two individuals who depend on each other on different networks have the same coupling intensity, which is called symmetric coupling; conversely, they are called asymmetric coupling. The computer simulation results show that the heterogeneous coupling between the networks can significantly improve the overall cooperation level of the network, and the symmetric coupling mechanism can promote the spread of cooperative behavior between game players more than the asymmetric coupling mechanism.
【学位授予单位】：天津理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：O157.5;O225

【相似文献】