A note on a generalized network flow model for manufacturing process

Manufacturing network flow （MNF） is a generalized network model that overcomes the limitation of an ordinary network flow in modeling more complicated manufacturing scenarios, in particular the synthesis of different materials into one product and/or the distilling of one type of material into many different products. Though a network simplex method for solving a simplified version of MNF has been outlined in the literature, more research work is still needed to give a complete answer whether some classical duality and optimality results of the classical network flow problem can be extended in MNF. In this paper, we propose an algorithmic method for obtaining an initial basic feasible solution to start the existing network simplex algorithm, and present a network-based approach to checking the dual feasibility conditions. These results are an extension of those of the ordinary network flow problem.     

Convergence of Stochastic Gradient Descent in Deep Neural Network

Stochastic gradient descent(SGD) is one of the most common optimization algorithms used in pattern recognition and machine learning.This algorithm and its variants are the preferred algorithm while optimizing parameters of deep neural network for their advantages of low storage space requirement and fast computation speed.Previous studies on convergence of these algorithms were based on some traditional assumptions in optimization problems.However,the deep neural network has its unique properties.Some assumptions are inappropriate in the actual optimization process of this kind of model.In this paper,we modify the assumptions to make them more consistent with the actual optimization process of deep neural network.Based on new assumptions,we studied the convergence and convergence rate of SGD and its two common variant algorithms.In addition,we carried out numerical experiments with LeNet-5,a common network framework,on the data set MNIST to verify the rationality of our assumptions.     

Minimizing maximum risk for fair network connection with interval data

In this paper we introduce a minimax model for network connection problems with interval parameters. We consider how to connect given nodes in a network with a path or a spanning tree under a given budget, where each link is associated with an interval and can be established at a cost of any value in the interval. The quality of an individual link （or the risk of link failure, etc.） depends on its construction cost and associated interval. To achieve fairness of the network connection, our model aims at the minimization of the maximum risk over all links used. We propose two algorithms that find optimal paths and spanning trees in polynomial time, respectively. The polynomial solvability indicates salient difference between our minimax model and the model of robust deviation criterion for network connection with interval data, which gives rise to NP-hard optimization problems.     

On 1-rate and 2-rate multicast 3-stage Clos networks

The 3-stage Clos network C（n, m, r） is considered as the most basic and popular multistage interconnection network which has been widely employed for data communications and parallel computing systems. Quite a lot of efforts has been put on the research of the 3- stage Clos network. Unfortunately, very little is known for the multirate multicast Clos network which is the most complicated case. Firstly a sufficient condition for 1-rate multicast networks to be SNB is given, from which a result for 2-rate multicast networks to be WSNB can easily be gotten. Furthermore, by using a reservation-scheme routing, more specific result for 2-rate multicast networks to be WSNB can be obtained for the case of one of them exceeding 1/2.     

An application of Bernstein-Durrmeyer operators

In the present paper, we find that the Bemstein-Durrmeyer operators, besides their better applications in approximation theory and some other fields, are good tools in constructing translation network. With the help of the de la Vallee properties of the Bernstein-Durrmeyer operators a sequence of translation network operators is constructed and its degree of approximation is dealt.     

THE STABILITY IN NEURAL NETWORK WITH DELAY AND DYNAMICAL THRESHOLDEFFECTS

1 IntroductionThe researches on dynamics of neural network systems have been arousinga great deal of interests for more than thirty years. In last ten years, many.researches focused on the study of dynamics of neural network models withdelay. In fact, neural networks often have times delay in reality, for exampledue to the finite switching speed of amplifiers in electronic neural networks.Thus, a large amount of excellent results appeared consecutively [1]--[5]. In thispaper, what we study is…     

Exponential Synchronization of the Linearly Coupled Dynamical Networks with Delays

In this paper,the authors investigate the synchronization of an array of linearly coupled identical dynamical systems with a delayed coupling.Here the coupling matrix can be asymmetric and reducible.Some criteria ensuring delay-independent and delay- dependent global synchronization are derived respectively.It is shown that if the coupling delay is less than a positive threshold,then the coupled network will be synchronized.On the other hand,with the increase of coupling delay,the synchronization stability of the network will be restrained,even eventually de-synchronized.     

Existence and Exponential Stability of Almost Periodic Solution for Hopfield Neural Network Equations with Almost Periodic Imput

By constructing Liapunov functions and building a new inequality, we obtain two kinds of sufficient conditions for the existence and global exponential stability of almost periodic solution for a Hopfield-type neural networks subject to almost periodic external stimuli. In this paper, we assume that the network parameters vary almost periodically with time and we incorporate variable delays in the processing part of the network architectures.     

Analysis of SE τ IR ω S epidemic disease models with vertical transmission in complex networks

When the role of network topology is taken into consideration, one of the objectives is to understand the possible implications of topological structure on epidemic models. As most real networks can be viewed as complex networks, we propose a new delayed SEτ IRωS epidemic disease model with vertical transmission in complex networks. By using a delayed ODE system, in a small-world (SW) network we prove that, under the condition R0 ≤ 1, the disease-free equilibrium (DFE) is globally stable. When R0 > 1, the endemic equilibrium is unique and the disease is uniformly persistent. We further obtain the condition of local stability of endemic equilibrium for R0 > 1. In a scale-free (SF) network we obtain the condition R1 > 1 under which the system will be of non-zero stationary prevalence.     

THE STABILITY OF A KIND OF DISCRETE-TIME HOPFIELD NEURAL NETWORKS WITH ASYMPTOTICAL WEIGHTED MATRICES

In this paper, the authors analyze the stability of a kind of discrete-time Hopfield neural network with asymptotical weighted matrix, which can be expressed as the product of a positive definite diagonal matrix and a symmetric matrix. We obtain that it has asymptotically stable equilibriums if the network is updated asynchronously, and asymptotically stable equilibriums or vibrating final stage with 2 period if updated synchronously. To prove these, Lassale's invariance principle in difference equation is applied.     

Design and implementation of a hybrid MIPv6/PMIPv6-based mobility management architecture

As two of the most promising candidate solutions for realizing the next-generation all-IP mobile Internet, Mobile IPv6 (MIPv6) is a host-based protocol supporting global mobility, while Proxy Mobile IPv6 (PMIPv6) is a network-based protocol supporting localized mobility. In order to take full advantage of both and enhance the mobility performance, a hybrid MIPv6/PMIPv6-based mobility management architecture is proposed in this paper. First, the optimized coexistence architecture of MIPv6 and PMIPv6 is presented. Based on this architecture, the Hybrid scheme is proposed, in which localized mobility and global mobility are handled by PMIPv6 and MIPv6 respectively to improve the efficiency. Then we propose the Hybrid+ scheme based on the Hybrid scheme. The Hybrid+ scheme incorporates a protocol selection algorithm, which takes into account the mobility characteristics of mobile nodes (MN) and network conditions. This allows it to select the most suitable mobility supporting protocol between the basic MIPv6 and the Hybrid scheme. Performance analysis using a two-layer hierarchical network model reveals that the Hybrid scheme reduces the signaling costs by more than 20% compared to the basic MIPv6 and Hierarchical Mobile IPv6 (HMIPv6). In addition, we have implemented the proposed mobility management architecture in a test-bed. The experimental results show that our Hybrid scheme can improve the handover performance of UDP and TCP sessions over the other mobility management protocols. Moreover, when the protocol selection algorithm is adopted as in the Hybrid+ scheme, the performance can be further improved by more than 50% under various scenarios.     

任播网络录用概率的平稳分析

任播服务在IPv6[1]中定义为一种标准的网络服务，它获得了越来越广泛的应用，已成为一种十分重要的网络服务．其研究虽已获得不少结果。但有关其服务质量之严格的数学分析并不多见．本文对[2，3][所建立的任播网络系统(ED，1)。(WD／D H，1)，和(WD／D B，1)进行了平稳分析．给出了这些系统在平稳条件下录用概率和拒绝概率的解析表达式．     

创新网络合作密度影响因素及其仿真分析

合作创新模式使得创新主体之间构成具有拓扑结构特性的创新网络,创新主体合作行为会影响创新网络的可持续发展。为探究不同影响因素作用下创新网络合作密度的变化,基于演化博弈理论和BA无标度网络理论,运用仿真方法,分析了网络规模、聚类系数、偏好性仿真、利益分配对创新网络合作密度的影响。结果表明:小规模的创新网络具有较低的聚类系数有利于网络合作密度的提升,大规模的创新网络具有较高的聚类系数有利于网络合作密度的提升;无论是大规模还是小规模创新网络,适度的偏好性模仿均能促进网络合作密度的提升;对于利益分配而言,无论是大规模还是小规模的创新网络,按劳分配是提升网络合作密度的最佳利益分配形式。     

双重网络嵌入对公司IPO后市场表现的影响研究

将社会网络嵌入理论从单一网络拓展至双重网络,分析上市公司同时嵌入风险投资网络和承销商网络对公司IPO上市后市场表现的影响。利用2004～2017年有风险投资支持的IPO公司数据,实证检验结果表明：(1)单独嵌入风险投资网络中心位置、单独嵌入承销商网络中心位置,均会提高公司IPO后市场表现;(2)同时嵌入风险投资网络中心位置和承销商网络中心位置,对公司IPO后短期市场表现的正向影响会相互替代;(3)同时嵌入风险投资网络中心位置和承销商网络中心位置,对公司IPO后长期市场表现的正向影响会互相补充。     

利益驱动对创新网络合作行为演化的影响机理及仿真——基于复杂网络拓扑结构视角

随着合作创新的网络化发展,创新网络中各主体的利益诉求对网络演化的影响不断加深。基于不同网络拓扑结构视角下不同利益对创新网络主体合作行为的影响,建立了创新网络演化博弈模型。在无标度与小世界网络拓扑结构下,结合声誉机制、学习行为、网络环境等因素,对创新网络中合作行为的演化进行了仿真研究。结果表明:创新网络的合作行为演化受到网络结构及规模的双重影响,在无标度网络下,合作率更加稳定;合作利益的公平分配促进创新网络合作行为的演化;机会利益的诱惑导致创新网络合作行为的显著波动。     

BP模式识别与Hopfield联想记忆识别能力的比较

神经网络可以进行模式识别,这一功能具有很高的实用价值.主要研究BP神经网络和离散Hopfield神经网络的识别能力.BP网络是前向神经网络,使用的是监督式学习规则.离散Hopfield是一种反馈网络,结构上输入、输出节点相连使得网络系统的输出状态具有动态变化特性.就数字,字母,手写体数字比较了BP网络和离散Hopfield网络的识别能力,得到了可观的结论,对两个网络的应用做更进一步的阐述.     

网络图修改方法的两个创新

随着社会的不断向前发展，网络图的应用越来越广泛。比如项目管理中的网络图、电网中的控制图和交通运输中的控制图等都是常见的网络图。网络图越复杂，修改起来越麻烦，本提出的网络图修改方法的两个创新能大大提高网络图修改方法的效率。网络图修改的第一个创新是网络图节点的高效删除．第二个创新是平行工序变为顺序工序的自动实现。     

A Network Algorithm for Realizing Constructivelythe Farkas Lemma Arising from the Quasi-Generalized Network

1.IntroductionLetRIbethespaceofallrealvectorsindexedbyelementsofafiniteseti,icachosenelemelltofi,SandS*apairofcompletelyorthogonalsubspacesofRI.Foragivenpartition(PI,P2,P3,P4)ofI--{ic}(i.e.PIuPZuP3uP4=I--{eo},andPinPj=acfori/j),letF={axled6S,Ax'.>0,Ax.20fore6PI,Ax.50foreEP2,Ax.~0foreEP3},F*={dyIAyES*,ac.>0,ace30foreEPI,ac50foreEP2,ac.~0foreEP4}.Now,theFarkasLemmacanbegenerallydescribedas[1]:oneandonlyoneofthefollowingtwostatemelltsholds:(i)ThereexistsaaxEF.(n)Thereexistsaac…     

我国股市网络稳定性及其宏观影响因素——基于三种网络拓扑结构、鲁棒性的实证研究

基于金融时间序列的多重分形特征及衡量市场风险的VaR模型,建立我国沪深股市的股票关联网络,实证研究三种网络拓扑结构特征,并使用协整检验方法分析网络稳定性和宏观经济变量间的长期均衡关系。结果表明:股票价格网络不具有无标度性,多标度网络和风险网络都具有无标度性;在三种网络中,风险网络具有更强的鲁棒性。此外,股市波动率和网络稳定系数间互为格兰杰因果关系,股市波动的前期变化能有效解释网络稳定性系数的变化;网络稳定性与宏观经济变量间具有长期的均衡关系,GDP增长率、消费者物价水平CPI对网络稳定性具有正向效应,利率对网络稳定性具有负向效应。风险网络的提出有助于分析我国股市的短期风险及稳定性,并为制定系统风险防御策略提供参考。     

基于时差分析的时标网络图探究

鉴于已有时差概念并不能充分反映CPM网络紧前和紧后工序时差的内在联系，文中引入三个新时差（前共后单时差，前单后共时差，双共时差）概念。针对目前工程项目管理中通用的时标网络图一早时标网络图和迟时标网络图，在路长定理的基础上，给出其理论依据，并在时差分析的基础上，随机绘出时标网络图，三个新时差的引入将有助于今后时标网络图的绘制，使网络技术的应用更具灵活性和实用性。