复杂网络上的意见动力学对谣言传播的影响

通过在SIR(susceptible-infected-recovered)模型中引入抑制者对谣言的辟谣机制研究了在线社交网络上的意见动力学对谣言传播的影响.在这一模型中,节点可以与自身的邻居组成1个群,传播者可以通过该群传播信息,抑制者也可以在此群中对信息发表意见进行辟谣.辟谣机制在降低未知者对于谣言的接受概率的同时也可以促使传播者向抑制者转变.本文采用ER(Erd?s-Rényi)随机网络、无标度网络以及真实的社交网络研究了抑制者的沉默概率对于谣言传播范围的影响.首先发现,谣言传播的过程以传播者的峰值为界可以分为两个阶段,即谣言自由传播的前期以及抑制者和传播者互相制衡的后期;其次,谣言的传播会随着抑制者的沉默概率的增大而突然暴发.在谣言暴发阈值之下,沉默概率的增大不会导致谣言传播范围显著增大,这是由于未知者在感知到谣言并转变为传播者后又迅速转变为抑制者;而当沉默概率达到谣言暴发阈值时,抑制者将不能控制传播者对谣言的传播从而导致抑制者的降低和谣言的暴发;最后,无标度上的谣言自由传播的前期阶段比随机网络持续的时间更短,从而使无标度上的谣言更难以暴发.本文的模型综合考虑了意见动力学和谣言传播的相互作用,更加真实地模拟了真实世界社交网络中的谣言传播过程.为谣言传播的控制和干预提供了一些有用的思路和见解.     

多层网络级联失效的预防和恢复策略概述

现实生活中,与国计民生密切相关的基础设施网络大多不是独立存在的,而是彼此之间相互联系或依赖的,于是用于研究这些系统的多层网络模型随之产生.多层网络中的节点在失效或者遭受攻击后会因"层内"和"层间"的相互作用而产生级联效应,从而使得失效能够在网络层内和层间反复传播并使得失效规模逐步放大.因此,多层网络比单个网络更加脆弱.多层网络级联失效产生的影响和损失往往是非常巨大的,所以对多层网络级联失效的预防和恢复的研究具有重大意义.就多层网络级联失效的预防而言,主要包含故障检测,保护重要节点,改变网络耦合机制和节点备份等策略.就多层网络发生级联失效后的恢复策略而言,主要包含共同边界节点恢复、空闲连边恢复、加边恢复、重要节点优先恢复、更改拓扑结构、局域攻击修复、自适应边修复等策略.     

Effects of Inertia on Evolutionary Prisoner's Dilemma Game

Considering the inertia of individuals in real life,we propose a modified Fermi updating rule,where the inertia of players is introduced into evolutionary prisoner's dilemma game(PDG) on square lattices.We mainly focus on how the inertia affects the cooperative behavior of the system.Interestingly,we find that the cooperation level has a nonmonotonic dependence on the inertia:with small inertia,cooperators will soon be invaded by defectors;with large inertia,players are unwilling to change their strategies and the cooperation level remains the same as the initial state;while a moderate inertia can induce the highest cooperation level.Moreover,effects of environmental noise and individual inertia are studied.Our work may be helpful in understanding the emergence and persistence of cooperation in nature and society.     

A 2-stage strategy updating rule promotes cooperation in the prisoner's dilemma game

In this study,we propose a spatial prisoner’s dilemma game model with a 2-stage strategy updating rule,and focus on the cooperation behavior of the system.In the first stage,i.e.,the pre-learning stage,a focal player decides whether to update his strategy according to the pre-learning factor β and the payoff difference between himself and the average of his neighbors.If the player makes up his mind to update,he enters into the second stage,i.e.,the learning stage,and adopts a strategy of a randomly selected neighbor according to the standard Fermi updating rule.The simulation results show that the cooperation level has a non-trivial dependence on the pre-learning factor.Generally,the cooperation frequency decreases as the pre-learning factor increases;but a high cooperation level can be obtained in the intermediate region of 3 < β < 1.We then give some explanations via studying the co-action of pre-learning and learning.Our results may sharpen the understanding of the influence of the strategy updating rule on evolutionary games.     

Evolutionary Prisoner＇s Dilemma Game Based on Division of Work

We propose a new two-type-player prisoner＇s dilemma game based on the division of work on a square lattice, in which a fraction of the population μ are assigned type A and the rest B. In a one-shot two-player game, we let both of their original payoffs be scaled by a same multiplicative factor α 〉 1, if two neighboring players are of different types; however we leave the payoffs unchanged if they are of the same type. Then we show that combined with the two-type setup, the square lattice can assist to induce different social ranks according to players＇ abilities to collect payoffs. Simulation results show that the density of cooperation is significantly promoted for a wide range of the temptation to defection parameters and that there are optimal values for both α and μ leading to the maximal cooperation level. We reach these results by analyzing the distribution of the players in the social ranks and we also show some typical snapshots of the system.     

Accelerating Consensus by Preferential Words in the Naming Game

The naming game is a model of nonequilibrium dynamics for the self-organized emergence of a language or a communication system. We study a modified version of the minimal naming game in which the speaker selects a word from its inventory with a probability proportional to exp（Rs ＊ α）, where Rs is the success ratio of the name and α is a tunable parameter. By investigating the effects of α on the evolutionary processes for both square lattice and scale-free networks, we find that the convergence time decreases with the increasing α on both two networks, which indicates that preferential selection of successful words can accelerate the reaching of consensus. More interestingly, for α 〉 0, we find that the relation between convergence time and α exhibits a power-law form.     

具有弱依赖组的复杂网络上的级联失效

复杂系统的功能通过节点之间的连接而维持,部分节点的失效会对系统的连通性造成破坏而影响整体的功能.除此之外,部分节点还会通过隐含依赖性而形成“依赖组”,其中一个节点的失效会对依赖组中其余节点造成损害.本文研究了“依赖组”的平均规模,规模分布和节点之间的依赖强度对网络级联失效和鲁棒性的影响.通过数值模拟和理论分析发现,网络的级联失效在尺度层次上可以分为“组内级联”和“组间级联”两个过程.在组内级联过程中,一个节点的失效会通过节点之间的依赖性对组内其他节点造成破坏,从而诱发更多节点的失效,进而产生更大的破坏力.在组间级联过程中,失效节点会引起网络发生破碎而导致组外节点脱离网络巨分支而失效,这就引起了失效节点的跨依赖组传播.在这两种失效过程的共同作用下,网络在级联失效过程中会表现出连续和不连续的两种渗流相变现象,这两种相变现象的发生与节点间的依赖强度、网络度分布以及依赖组规模分布有关.这意味着通过控制依赖组的特征,如依赖组中节点之间的依赖强度或依赖组规模分布,可以避免系统突然崩溃进而提高网络的鲁棒性.     

Effect of Geometric Distance on Agreement Dynamics of Naming Game

We investigate the naming game on geometric networks. The geometric networks are constructed by adding geometric links to two-dimensional regular lattices. It is found that the agreement time is a non-monotonic function of the geometric distance and there exists an optimal value of the geometric distance resulting in the shortest agreement time. All these results show that the geometric distance plays an important role in the evolutionary process of the language game. Our results also show that the convergence time strongly depends on the number of adding links.