基于人工神经网络的商业银行信用风险模型

在对人工神经网络的基本原理进行简要介绍的基础上 ,着重对构建商业银行信用风险的人工神经网络模型进行了研究 ,实证结果表明 ,人工神经网模型具有很高的预测精度     

资源有限网络计划的PRWI启发式优化方法

本文在综合考虑了有资源约束的网络计划结构特征、资源强度、时间约束等方面因素的基础上，提出了一种新的资源优化的启发式优化方法—ＰＲＷＩ方法，并通过分析证明了该方法处理问题的效果较现有的其它方法好。     

Use of dynamic trees in a network simplex algorithm for the maximum flow problem

Goldfarb and Hao (1990) have proposed a pivot rule for the primal network simplex algorithm that will solve a maximum flow problem on ann-vertex,m-arc network in at mostnm pivots and O(n ² m) time. In this paper we describe how to extend the dynamic tree data structure of Sleator and Tarjan (1983, 1985) to reduce the running time of this algorithm to O(nm logn). This bound is less than a logarithmic factor larger than those of the fastest known algorithms for the problem. Our extension of dynamic trees is interesting in its own right and may well have additional applications.Research partially supported by a Presidential Young Investigator Award from the National Science Foundation, Grant No. CCR-8858097, an IBM Faculty Development Award, and AT&T Bell Laboratories.Research partially supported by the Office of Naval Research, Contract No. N00014-87-K-0467.Research partially supported by the National Science Foundation, Grant No. DCR-8605961, and the Office of Naval Research, Contract No. N00014-87-K-0467.     

用于光学神经网络硬件系统误差纠正的虚拟神经网络

提出了一种用于修正光学神经网络硬件系统误差的虚拟神经网络模型，光学实验结果表明该网络能够有效地消除硬件系统误差对实验结果的影响。     

IPv4/IPv6翻译网关的研究

In this paper, a model of translation gateway is proposed. The communications between IPv4 network and IPv6 network are realized by using the Microsoft intermediate driver technology in environment of Windows 2000.     

Stability criteria for controlled queueing systems

We give an almost complete classification of ergodicity and transience conditions for a general multi-queue system with the following features: arrivals form Poisson streams and there are various routing schemes for allocating arrivals to queues; the servers can be configured in a variety of ways; completed jobs can feed back into the system; the exponential service times and feedback probabilities depend upon the configuration of the servers (this model includes some types of multi-class queueing system); switching between service regimes is instantaneous. Several different levels of control of the service regimes are considered. Our results for the N-queue system require randomisation of service configurations but we have studied the two queue system in situations where there is less control. We use the semi-martingale methods described in Fayolle, Malyshev and Menshikov [3] and our results generalise Kurkova [8] and complement Foley and McDonald [4] and [5]. AMS 2000 subject classification: Primary: 90B22; Secondary: 60J10 90B15     

Locally minimal uniformly oriented shortest networks

The Steiner problem in a λ-plane is the problem of constructing a minimum length network interconnecting a given set of nodes (called terminals), with the constraint that all line segments in the network have slopes chosen from λ uniform orientations in the plane. This network is referred to as a minimum λ-tree. The problem is a generalization of the classical Euclidean and rectilinear Steiner tree problems, with important applications to VLSI wiring design.A λ-tree is said to be locally minimal if its length cannot be reduced by small perturbations of its Steiner points. In this paper we prove that a λ-tree is locally minimal if and only if the length of each path in the tree cannot be reduced under a special parallel perturbation on paths known as a shift. This proves a conjecture on necessary and sufficient conditions for locally minimal λ-trees raised in [M. Brazil, D.A. Thomas, J.F. Weng, Forbidden subpaths for Steiner minimum networks in uniform orientation metrics, Networks 39 (2002) 186-222]. For any path P in a λ-tree T, we then find a simple condition, based on the sum of all angles on one side of P, to determine whether a shift on P reduces, preserves, or increases the length of T. This result improves on our previous forbidden paths results in [M. Brazil, D.A. Thomas, J.F. Weng, Forbidden subpaths for Steiner minimum networks in uniform orientation metrics, Networks 39 (2002) 186-222].     

Electro-optical performance of polymer stabilized cholesteric texture cell: The influence of chiral dopant and monomer concentration

Reverse-mode polymer stabilized cholesteric texture (PSCT) cells with various chiral dopant and monomer concentrations were fabricated. Experimental results indicate that the chiral dopant concentration and the monomer concentration significantly influence the threshold voltage and hysteresis of the cells. The chiral dopant and monomer deform the hysteresis loop. Increasing the monomer concentration reduces the width of the hysteresis loop of the reverse-mode PSCT cell.     

概率神经网络和FTIR光谱用于食道癌的辅助分析

利用正常与相应癌化食道组织的主要FTIR特征峰aυs,CH3、sυ,CH2、σCH2、aυs,po4-、υc-o、sυ,po2-及sυ,磷酸化蛋白作为概率神经网络的输入向量,对网络的主要参数(网络径向基函数分布spread(0~5))、输入向量和网络表现(m ean accurate rate of recogn ition)之间的关系进行了研究。主要结论如下:i)无论输入向量是哪种特征频率的组合,其平均识别正确率都高于71.40%;ii)当输入向量为特征频率sυ,po2、sυ,磷酸化蛋白或υc-0、sυ,po2、sυ,磷酸化蛋白时,网络表现较佳,平均识别正确率较好。当spread介于1.4~2.3时,两者均达到网络具有的最高平均识别正确率(85.71%);iii)大多数情况下,网络的平均识别正确率与spread之间呈现二个高峰的特征,即spread介于0.1~0.3和1.5~5.0之间时,网络均具有较高的平均识别正确率。研究表明,以傅里叶变换红外光谱的主要特征峰为概率神经网络的输入向量,用于食道组织样品的癌化识别分析是完全可能的,其平均识别正确率可达85.71%。     

线性相关计算法形成数字剪切散斑相关条纹图

本文提出了用线性相关计算法形成数字剪切散斑干涉条纹的新方法,从理论分析和实验研究均证明,用该方法所获得的数字剪切散斑相关条纹图在图象监视器上易于观看,条纹图象光强无直流分量影响且可见度较减模式要高,该新方法的采用为今后的条纹图象分析奠定了基础。