复杂网络统计力学的差分方程方法

通过分析几种估计增长网络度分布方法的缺点,提出估计度分布的差分方程方法,不仅避免了复杂网络分析中将离散问题连续化带来的逻辑矛盾,也避免了网络稳态度分布存在性的假设.利用这个方法给出Poisson增长择优连接网络的度分布公式,借助Poisson过程理论和Gamma 分布的性质严格证明Poisson增长择优连接网络是无标度网络.     

随机增长网络的度分布

增长和择优机制是无标度网络中的两种重要的演化机制,在分析BA模型的基础上,提出了一种新的节点增长方式,即考虑了新增节点的连边数是随机变量的情况,从而建立了随机增长网络模型,并利用随机过程理论得到了在这种增长方式下网络的度分布,结果表明这个网络是无标度网络。     

基于二项分布随机增长的无标度网络

提出—个具有随机增长的无标度网络模型．该模型的演化规则仍然是BA模型的增长和择优连接,但是每一时间间隔添加到网络中的边数是—个具有二项分布的随机变量．通过率方程方法,本文证明了该网络的度分布具有幂律尾部,该模型生成了—个无标度网络．     

M-G-P型复杂网络中有寿命节点的进入退出机制

现实中复杂网络结构复杂,形式多样,处在高度动态变化的过程.为了更好地理解真实网络的演化,基于复杂网络的特性进行分析,建立了Poissotn连续时间增长节点具有寿命的M-G-P型复杂网络模型,模型中包括:新节点加入、节点老化和老节点退出等,基于齐次马尔可夫链对模型的度分布进行计算,得出M-G-P型网络的度分布符合幂律分布,模型和BA模型一样能产生指数γ=3的无标度网络,验证了导致无标度网络度分布特征起关键性作用的是链接的偏好特性.     

基于贝叶斯网络的工程项目进度完工概率分析

针对传统计划评审技术(Program Evaluation and Review Technique,PERT)在计算完工概率时假设条件的局限性(假设条件与工程实际存在偏差,导致完工概率偏大),提出了基于贝叶斯网络的施工进度完工概率分析方法.首先,分析了贝叶斯网络与进度计划网络之间的相似性,将两者结合起来构建了贝叶斯进度网络;在此基础上,综合考虑贝叶斯网络在节点取值及概率计算方面的优越性,并结合工程项目的不确定性及复杂性特点,建立了基于贝叶斯网络的施工进度完工概率分析模型.最后,将该模型应用于具体工程进行实例分析,验证了模型的可行性与有效性.研究结果表明:基于贝叶斯网络的进度完工概率模型充分考虑了工程施工中的风险因素,其结果能更客观地反映工程实际,可为工程项目决策者提供可靠的依据.     

一类增长网络模型的度分布

将主方程方法和马氏链首达概率方法应用于一个去边机制与时间相关的网络模型,得到这个模型度分布的精确表达式,并严格证明了度分布的存在性.     

一类推广的双险种复合Poisson风险模型的破产概率

本文研究了一类索赔计数过程相关的双险种Poisson风险模型.利用模型转化首先将该复杂模型转化为经典的风险模型,获得该模型破产概率所满足的积分方程,Lundberg上界表达式,及Cramér-Lundberg渐近估计式.当个体索赔具有指数分布时,推得了破产概率所满足的方程,并给出了具体的数值计算的实例.     

随机和择优混合演化网络模型的稳定性

本文运用主方程等方法给出随机和择优混合演化网络模型稳态度分布存在性的严格证明,并推导度分布的表达式,进而得知该混合演化网络为无标度网络.     

一类混合增长网络的度分布

本文提出了一种新的带择优的混合增长网络模型,并用马氏链理论严格证明了其稳定度分布的存在性。     

具有随机需求的供应链网络缺货概率的计算方法

由于需求的不确定性及网络的复杂性,使得供应链网络上各企业缺货概率的计算变得非常棘手.将供应链网络模型成Markov过程,利用排队理论提出了供应链网络各企业缺货及因满货而待送货的稳态概率的计算公式;对于较大的供应链网络,提出了将系统分解成2级供应链网络集合求各企业缺货概率的近似方法.这种方法是通过修正各制造商的批量送货间隔时间、各销售商及各零售商的批量销售间隔时间参数,把分解后的子系统与多级供应链网络连接起来,利用子系统求出各企业的缺货和因满货而待送货的稳态概率的近似值.数值试验表明所提出的近似解法具有很高的精度.     

A preferential attachment model with Poisson growth for scale-free networks

We propose a scale-free network model with a tunable power-law exponent. The Poisson growth model, as we call it, is an offshoot of the celebrated model of Barabási and Albert where a network is generated iteratively from a small seed network; at each step a node is added together with a number of incident edges preferentially attached to nodes already in the network. A key feature of our model is that the number of edges added at each step is a random variable with Poisson distribution, and, unlike the Barabási–Albert model where this quantity is fixed, it can generate any network. Our model is motivated by an application in Bayesian inference implemented as Markov chain Monte Carlo to estimate a network; for this purpose, we also give a formula for the probability of a network under our model.     

An individual-based modeling framework for infectious disease spreading in clustered complex networks

We introduce an individual-based model with dynamical equations for susceptible-infected-susceptible (SIS) epidemics on clustered networks. Linking the mean-field and quenched mean-field models, a general method for deriving a cluster approximation for three-node loops in complex networks is proposed. The underlying epidemic threshold condition is derived by using the quasi-static approximation. Our method thus extends the pair quenched mean-field (pQMF) approach for SIS disease spreading in unclustered networks to the scenario of epidemic outbreaks in clustered systems with abundant transitive relationships.We found that clustering can significantly alter the epidemic threshold, depending nontrivially on topological details of the underlying population structure. The validity of our method is verified through the existence of bounded solutions to the clustered pQMF model equations, and is further attested via stochastic simulations on homogeneous small-world artificial networks and growing scale-free synthetic networks with tunable clustering, as well as on real-world complex networked systems. Our method has vital implications for the future policy development and implementation of intervention measures in highly clustered networks, especially in the early stages of an epidemic in which clustering can decisively alter the growth of a contagious outbreak.     

探讨几类具有完全图核心的网络模型的优美性

在复杂网络研究中,人们需要建立网络模型,无标度图就是这样的一种网络模型.我们发现具有完全图核心的网络模型可以演变成无标度图.具有完全图核心的几种网络模型的优美性得到研究.     

EXACT SOLUTION OF THE DEGREE DISTRIBUTION FOR AN EVOLVING NETWORK

In this paper we propose a simple evolving network with link additions as well as removals. The preferential attachment of link additions is similar to BA model’s, while the removal rule is newly added. From the perspective of Markov chain, we give the exact solution of the degree distribution and show that whether the network is scale-free or not depends on the parameter m, and the degree exponent varying in （3, 5] is also depend on m if scale-free.     

Poisson增长竞争网络与适应度模型

提出吸引度依赖于时间的竞争网络模型.利用Poisson过程获得这个模型稳态平均度分布的解析表达式.理论分析表明,这类网络幂律指数与渐近吸引系数和新节点边数m有关,且在区间(1+1/m,m+1)内.作为竞争网络模型的应用,获得了适应度模型的度分布估计.结果表明适应度模型是竞争网络模型的特例,反之则不然.     

Risk-averse dynamic programming for Markov decision processes

We introduce the concept of a Markov risk measure and we use it to formulate risk-averse control problems for two Markov decision models: a finite horizon model and a discounted infinite horizon model. For both models we derive risk-averse dynamic programming equations and a value iteration method. For the infinite horizon problem we develop a risk-averse policy iteration method and we prove its convergence. We also propose a version of the Newton method to solve a nonsmooth equation arising in the policy iteration method and we prove its global convergence. Finally, we discuss relations to min–max Markov decision models.     

BA模型的数学基础——纪念李国平院士吴新谋教授诞辰100周年

该文基于马氏链的概念和技巧, 给出了BA无标度网络模型稳态度分布存在性的严格证明, 并且从数学上重新推导了度分布的精确解析表达式. 此处所用的方法具有一定的普适性, 适用于更一般的无标度型复杂网络模型.     

Generating hierarchial scale-free graphs from fractals

Motivated by the hierarchial network model of E. Ravasz, A.-L. Barabási, and T. Vicsek, we introduce deterministic scale-free networks derived from a graph directed self-similar fractal Λ. With rigorous mathematical results we verify that our model captures some of the most important features of many real networks: the scale-free and the high clustering properties. We also prove that the diameter is the logarithm of the size of the system. We point out a connection between the power law exponent of the degree distribution and some intrinsic geometric measure theoretical properties of the underlying fractal. Using our (deterministic) fractal Λ we generate random graph sequence sharing similar properties.