An Approach to Modeling Artificial Gene Networks

We propose a new mathematical model of a repressilator, i.e., the simplest gene ring network consisting of three elements. The studied model is a three-dimensional system of ordinary differential equations depending on a single parameter. We study the existence and stability problems for relaxation periodic motion in this system.     

Artificial Intelligence Approaches to Simulation

Recent years have seen a number of attempts to apply both AI methods and AI software tools to the development of both simulations and simulation packages. This paper discusses the motivation behind such work: the increasing need for AI researchers to include simulation in AI systems and use simulation in research, and the need to model intelligent behaviour in OR simulation models. The current research and development effort is reviewed. A system called PROSS (the PROlog Simulation System), constructed to allow the integration of AI methods with traditional discrete simulation, is then briefly presented. It is concluded that AI software tools, and some of the novel approaches to simulation being taken by AI researchers, will have a considerable impact on simulation.     

An Approach to Improving Consistency of Fuzzy Preference Matrix

Based on the transfer formulas of fuzzy preference matrix and multiplicative preference matrix, this paper presents an approach to improving consistency of fuzzy preference matrix and gives a practical iterative algorithm to derive a modified fuzzy preference matrix with acceptable consistency. By using two criteria, we can judge whether the modification is acceptable or not. Finally, a numerical example is given to show the feasibility and effectiveness of the algorithm.     

三层前向人工神经网络全局最优逼近

提出了求解不等式约束非线性优化问题的群体复合形进化算法 ,提出的算法能充分利用目标函数值的信息、优化搜索过程具有较强的方向性和目标性 ,收敛速度较快 ,且是全局优化算法 ;将群体复合形进化算法应用于三层前向人工神经网络逼近 ,提出了三层前向人工神经网络全局最优逼近算法 ;将三层前向人工神经网络全局最优逼近算法应用于实例 ,表明了提出的全局最优逼近算法的有效性 .     

An Anthology on the Homology of Simulation with Artificial Intelligence

Simulation and artificial intelligence (AI) can be related in the following ways: the methodological similarities, the actual and potential uses of AI concepts in simulation, expert systems in simulation environments, and the gains for AI when applying ideas from simulation. This paper highlights the personal experiences of the author by discussing, for each of the four above-mentioned areas, a topic or a computer system in which he is directly involved. To highlight the similarities, it will be shown that the three-phase simulation model can be considered as a production-system model. A natural-language understanding system that was used in the definition stage of a simulation project will provide the basis for discussing AI concepts in simulation environments. An expert system used in the development and verification stages will provide an insight into the application of expert systems in simulation modelling. An expert-system development tool which is partially based on simulation techniques will be used to outline the gains for AI.     

处理网络计划的不确定性的一种进度模型

在应用网络计划技术进行工程项目管理中,常常需要对工程项目中潜在的不确定性进行处理,而传统的网络计划技术中,缺乏对工程项目中潜在的不确定性的处理。章通过建立评估不确定性的影响因子区间的模型与方法,提出了在工程项目的进度管理中动态地对潜在的不确定性进行处理的有效方法。     

Artificial Intelligence—An Overview of Similarities with O.R.

A brief outline of A.I. is provided, noting problems, techniques and objectives which are similar to those found in O.R. It is intended that this will indicate to O.R. workers which A.I. areas might be worth following up as relevant to their own work and which problems in A.I. might be attacked by the O.R. community as part of the development of intelligent support systems.     

An Efficient Approach to Real-Time Traffic Routeing for Telephone Network Management

When contingencies occur in the telephone network the proper traffic control actions for rearranging the traffic flow must be promptly taken in a short period of time, otherwise, the congestion will soon spread to other parts of the network. In this paper, a multicommodity maximal flow model is proposed to formulate the problem, and an efficient approach is proposed to find the solution within a short time period. The node pairs are first sorted by the traffic demand into decreasing order and the capacity of links is allocated to the node pair with largest demand. The traffic allocation is then iteratively reallocated on those routes which share the same links to find a better solution, until that solution is obtained. The advantages of this approach are: (1) a feasible solution can be obtained within a short time period; (2) the affected traffic can be rerouted on paths with more than two links; (3) an acceptable solution can be found once the period has expired.     

Computer Mediated Learning: An Example of an Approach

There are several possible approaches in which dynamic computerized environments play a significant, and possibly unique, role in supporting innovative learning trajectories in mathematics in general, and particularly in geometry. These approaches are influenced by the way one views mathematics and mathematical activity.In this paper we briefly describe an approach based on a problem situation and our experiences using it with students and teachers. This leads naturally to a discussion of some of the ways in which parts of the mathematics curriculum, classroom practice, and student learning may differ from the traditional approach.This revised version was published online in September 2005 with corrections to the Cover Date.     

利用人工智能案例推动概率统计课程教学

人工智能的迅猛发展离不开概率统计的支持,在概率统计的日常教学中,融入人工智能案例的介绍,能够促进学生的学习兴趣,提高教学质量.以举例的方式说明了概率统计的理论知识和人工智能技术的结合,为利用人工智能案例推动概率统计的课程教学提供了示例.     

An Approximation Method for Improving Dynamic Network Model Fitting

There has been a great deal of interest recently in the modeling and simulation of dynamic networks, that is, networks that change over time. One promising model is the separable temporal exponential-family random graph model (ERGM) of Krivitsky and Handcock, which treats the formation and dissolution of ties in parallel at each time step as independent ERGMs. However, the computational cost of fitting these models can be substantial, particularly for large, sparse networks. Fitting cross-sectional models for observations of a network at a single point in time, while still a nonnegligible computational burden, is much easier. This article examines model fitting when the available data consist of independent measures of cross-sectional network structure and the duration of relationships under the assumption of stationarity. We introduce a simple approximation to the dynamic parameters for sparse networks with relationships of moderate or long duration and show that the approximation method works best in precisely those cases where parameter estimation is most likely to fail—networks with very little change at each time step. We consider a variety of cases: Bernoulli formation and dissolution of ties, independent-tie formation and Bernoulli dissolution, independent-tie formation and dissolution, and dependent-tie formation models.     

基于人工神经网络的客户价值评价体系

企业对客户价值的评价是进行客户分类及实施客户策略的依据,也是实现客户关系管理的核心所在.客户为企业带来的价值本身是具有层次性和动态性,然基于客户全生命周期的客户终生价值计算方法显然无法将其动态发展过程准确的考虑在内,也就无法规避其计算结果所带来的误差风险.因此,在现有客户价值评价理论的基础上,结合客户生命周期理论,以企业与客户的双赢为最终目标,提出分阶段细化的评价思想,构建完善的评价体系,并根据该指标体系的指标特性选择人工神经网络作为评价方法,给出模型求解,最后依据此评价结果对客户分类及营销策略进行相关的分析.     

An Incremental Procedure for Improving Path Assignment in a Telecommunications Network

A fundamental problem in the design and management of a telecommunications network is that of determining an optimal routing pattern for a given set of origin-destination demand pairs. In addition, reliability considerations may require provisioning a set of backup paths to protect the working traffic against network failures. In the literature, the problem of finding an optimal routing for a network with fixed link capacities and a list of point-to-point demands (origin-destination pairs), each with a set of candidate routing paths has been referred to as the path-assignment problem. There are three versions of this problem that correspond to the type of network protection required (no protection, dedicated protection, and shared protection). The solution to those models can be used to determine an initial design for a new network. Over time, however, changes in the demand pattern and/or upgrades to the network equipment may create a situation in which the working and/or backup paths are sub-optimal.For network managers who are reluctant to make wholesale changes to an established and reliable routing assignment, a complete modification to obtain an optimal assignment that uses fewer network resources is viewed as highly risky. This investigation presents a new procedure to take a given feasible, but sub-optimal design and improve it by making a series of incremental improvements each of which only changes a small number of path assignments. Network managers view this strategy as much less risky since only a few customers are affected by any one change. Test cases that require no protection, dedicated protection, and shared protection were examined in an empirical analysis. For all cases, near-optimal solutions were achieved irrespective of the quality of the given sub-optimal starting solutions.     

Game-Theoretic Approach to the Optimization of FDDI Computer Networks1

The problem of maximizing the performance in a fiber distributed data interface (FDDI) computer network is formulated as a cooperative n-player game. Solutions of this game can be obtained by solving special optimization problems. Using the models and formulas developed by Tangemann (Ref. 1) and Klehmet (Ref. 2) for the mean waiting times, the resulting optimization problems are presented and numerical results are given.     

认股权证的保险精算定价

引入Mogens Bladt和Tina Hviid Rydberg在无市场假设下关于期权定价的保险精算方法,利用公平保费原则和价格过程的实际概率测度,建立认股权证的定价模型,并给出定价公式.当投资者对原生资产期望回报率为无风险利率时,该定价为风险中性价格.     

An Elementary Approach to Exponential Spaces

In 1970, Day and Kelly characterized exponential spaces by a condition (C). Eight years later, Hofmann and Lawson pointed out that this is equivalent to quasi-local compactness, i.e. every neighborhood V of a point contains a smaller one W such that any open cover of V admits a finite subcover of W. These characterizations work with topologies on topologies and may be felt to be not really elementary. This note instead offers an elementary approach which applies to quotient-reflective subcategories as well and includes a natural generalization of the compact-open topology on function spaces.     

An Algorithmic Approach to Chain Recurrence

In this paper we give a new definition of the chain recurrent set of a continuous map using finite spatial discretizations. This approach allows for an algorithmic construction of isolating blocks for the components of Morse decompositions which approximate the chain recurrent set arbitrarily closely as well as discrete approximations of Conley’s Lyapunov function. This is a natural framework in which to develop computational techniques for the analysis of qualitative dynamics including rigorous computer-assisted proofs.