现代优化计算方法在蛋白质结构预测中的应用

现代优化计算方法在蛋白质结构预测中占有重要地位.简要地介绍了模拟退火算法,遗传算法,人工神经网络和图论算法在蛋白质结构预测中的应用.对国内外近年来应用这些算法,特别是在蛋白质构象搜索问题中,解决蛋白质结构预测的研究作了回顾,并分析、比较了这几种算法的效果和特点.     

On the ill conditioning of locating transmission zeros in least squares ARMA filtering

The algorithms of Levinson-Schur and Nevanlinna-Pick are briefly reviewed. Both produce least squares predictive filters. By minimizing the least squares error with respect to the interpolation points of the Nevanlinna-Pick algorithm we find the transmission zeros of an ARMA filter. It is shown by some simple examples that this is an ill conditioned problem.     

Computational methods for aerodynamic shape design

In this paper, various computational methods for shape design in aerodynamics are reviewed. A classification system based on the mathematical structure of the methods is introduced. The fundamental ideas of different formulations are discussed, and the algorithms are described. The advantages and well-known disadvantages of different methods are also briefly discussed.     

The inefficiency of least squares: extensions of the kantorovich inequality

Four different measures of inefficiency of the simple least squares estimator in the general Gauss-Markoff model are considered. Previous work on the bounds to some of these measures is briefly reviewed, and new bounds are obtained for a particular measure.     

正交约束优化问题的一阶算法

带有正交约束的矩阵优化问题在材料计算、统计及数据分析等领域中有着广泛的应用.由于正交约束的可行域是Stiefel流形,一直以来流形上的优化方法是求解这一问题的主要方法.近年来,随着实际应用问题所要求的变量规模的扩大,传统的流形优化方法在计算上的劣势显现出来,而一些迭代简单、收敛快的新算法逐渐被提出.通过收缩方法、非收缩可行方法、不可行方法三个类别分别来介绍求解带有正交约束的矩阵优化问题的最新算法.通过分析这些方法的主要特性,以及应用问题的要求,对这类问题算法设计的研究进行了展望.     

经典一维装箱问题近似算法的研究进展

自20世纪70年代开始,随着计算复杂性理论的建立,近似算法逐渐成为组合优化的重要研究方向。作为第一批研究对象,装箱问题引起了组合优化领域学者的极大关注。装箱问题模型简单、拓展性强,广泛出现在各种带容量约束的资源分配问题中。除了在物流装载和材料切割等方面愈来愈重要的应用外,装箱算法的任何理论突破都关乎到整个组合优化领域的发展。直到今天,对装箱问题近似算法的研究仍如火如荼。本文主要针对一维模型,简述若干经典Fit算法的发展历程,分析基于线性规划松弛的近似方案的主要思路,总结当前的研究现状并对未来的研究提供一些参考建议。     

经典一维装箱问题近似算法的研究进展

自20世纪70年代开始,随着计算复杂性理论的建立,近似算法逐渐成为组合优化的重要研究方向。作为第一批研究对象,装箱问题引起了组合优化领域学者的极大关注。装箱问题模型简单、拓展性强,广泛出现在各种带容量约束的资源分配问题中。除了在物流装载和材料切割等方面愈来愈重要的应用外,装箱算法的任何理论突破都关乎到整个组合优化领域的发展。直到今天,对装箱问题近似算法的研究仍如火如荼。本文主要针对一维模型,简述若干经典Fit算法的发展历程,分析基于线性规划松弛的近似方案的主要思路,总结当前的研究现状并对未来的研究提供一些参考建议。     

Descriptive statistics for multivariate distributions

The purpose of this paper is to study the concepts location, scatter, skewness and kurtosis of multivariate distributions. Measures of these properties are introduced which include some new generalizations of well-known univariate statistics. Previous work is briefly reviewed.     

Algorithms for low-dimensional topology

In this article, we re-introduce the so called “Arkaden–Faden–Lage” (briefly, AFL) representation of knots in three-dimensional space introduced by Kurt Reidemeister in the 1930s and show how it can be used to develop efficient algorithms to compute some important topological knot structures. In particular, we introduce an efficient algorithm to calculate the holonomic representation of knots introduced by V. Vassiliev and give the main ideas on how to use the AFL representations of knots to compute the Kontsevich integral. The methods introduced here are to our knowledge novel and can open new perspectives in the development of fast algorithms in low-dimensional topology.     

Economy-Wide Models for Developing Countries,with Emphasis on Food Production and Distribution

Hunger and malnutrition are persistent problems in many developing countries. Policies required to overcome these problems are difficult to identify, whereas the experience in planning policies in their broadest sense for this purpose is limited. In this paper, economy-wide models constructed for this purpose by the Centre for World Food Studies are described, with special emphasis on a model for Bangladesh. Its application is demonstrated in assessing alternative policies for two specific cases, i.e. the subsidization of synthetic fertilizers and food rationing. Other applications are briefly described. Construction of the models has involved the development of software to facilitate the execution and organization of the work and also problem-specific algorithms. Reference is made to these in the paper.     

Probability-one homotopies in computational science

Probability-one homotopy algorithms are a class of methods for solving nonlinear systems of equations that, under mild assumptions, are globally convergent for a wide range of problems in science and engineering. Convergence theory, robust numerical algorithms, and production quality mathematical software exist for general nonlinear systems of equations, and special cases such as Brouwer fixed point problems, polynomial systems, and nonlinear constrained optimization. Using a sample of challenging scientific problems as motivation, some pertinent homotopy theory and algorithms are presented. The problems considered are analog circuit simulation (for nonlinear systems), reconfigurable space trusses (for polynomial systems), and fuel-optimal orbital rendezvous (for nonlinear constrained optimization). The mathematical software packages HOMPACK90 and POLSYS_PLP are also briefly described.     

Research topics in discrete estimation of distribution algorithms based on factorizations

In this paper, we identify a number of topics relevant for the improvement and development of discrete estimation of distribution algorithms. Focusing on the role of probability distributions and factorizations in estimation of distribution algorithms, we present a survey of current challenges where further research must provide answers that extend the potential and applicability of these algorithms. In each case we state the research topic and elaborate on the reasons that make it relevant for estimation of distribution algorithms. In some cases current work or possible alternatives for the solution of the problem are discussed.     

实规范矩阵的特征值问题

实对称矩阵的特征值问题,无论是低阶稠密矩阵的全部特征值问题,或高阶稀疏矩阵的部分特征值问题,都已有许多有效的计算方法,迄今最重要的一些成果已总结在[5]中。本文利用规范矩阵的一些重要性质将对于Hermite矩阵(特别是对弥矩阵)特征值问题的一些有效算法推广到规范矩阵的特征值问题,由于对复规范阵的推广是简单的,而且实际上常遇到的是实矩阵(这时常要求只用实运算),因此我们着重讨论实规范矩阵的特征值问题。     

Parallel computing in nonconvex programming

In this paper, we are concerned with the development of parallel algorithms for solving some classes of nonconvex optimization problems. We present an introductory survey of parallel algorithms that have been used to solve structured problems (partially separable, and large-scale block structured problems), and algorithms based on parallel local searches for solving general nonconvex problems. Indefinite quadratic programming posynomial optimization, and the general global concave minimization problem can be solved using these approaches. In addition, for the minimum concave cost network flow problem, we are going to present new parallel search algorithms for large-scale problems. Computational results of an efficient implementation on a multi-transputer system will be presented.     

Algorithms for linear programming with linear complementarity constraints

Linear programming with linear complementarity constraints (LPLCC) is an area of active research in Optimization, due to its many applications, algorithms, and theoretical existence results. In this paper, a number of formulations for important nonconvex optimization problems are first reviewed. The most relevant algorithms for computing a complementary feasible solution, a stationary point, and a global minimum for the LPLCC are also surveyed, together with some comments about their efficiency and efficacy in practice.     

Computational Reduction for Parametrized PDEs: Strategies and Applications

In this paper we present a compact review on the mostly used techniques for computational reduction in numerical approximation of partial differential equations. We highlight the common features of these techniques and provide a detailed presentation of the reduced basis method, focusing on greedy algorithms for the construction of the reduced spaces. An alternative family of reduction techniques based on surrogate response surface models is briefly recalled too. Then, a simple example dealing with inviscid flows is presented, showing the reliability of the reduced basis method and a comparison between this technique and some surrogate models.     

Machine learning applications in nonlife insurance

The literature on analytical applications in insurance tends to be either very general or rather technical, which may hold back the adoption of new important tools by industrial practitioners. Our goal is to stress that machine learning (ML) algorithms will play a significant role in the insurance industry in the near future and thus to encourage practitioners to learn and apply these techniques. After discussing the increasing relevance of data for nonlife insurance and briefly reviewing the major impact of digital technology on this business, we restrict our discussion to technical analytical applications and indicate where ML algorithms can add most value. We present two real examples: first a comparison of retention models for household insurance and then a dynamic pricing problem for online motor insurance. Both applications illustrate the advantages but also some of the difficulties of applying ML tools in practice. Finally, we mention some challenges posed by the use of ML in the industry and formulate a few recommendations for successful applications in insurance. This article is neither a tutorial nor an exhaustive review of technical ML applications in nonlife insurance. However, references for additional learning materials are provided.     

Approximation algorithms for nonnegative polynomial optimization problems over unit spheres

We consider approximation algorithms for nonnegative polynomial optimization problems over unit spheres. These optimization problems have wide applications e.g., in signal and image processing, high order statistics, and computer vision. Since these problems are NP-hard, we are interested in studying on approximation algorithms. In particular, we propose some polynomial-time approximation algorithms with new approximation bounds. In addition, based on these approximation algorithms, some efficient algorithms are presented and numerical results are reported to show the efficiency of our proposed algorithms.     

A survey of simulated annealing as a tool for single and multiobjective optimization

This paper presents a comprehensive review of simulated annealing (SA)-based optimization algorithms. SA-based algorithms solve single and multiobjective optimization problems, where a desired global minimum/maximum is hidden among many local minima/maxima. Three single objective optimization algorithms (SA, SA with tabu search and CSA) and five multiobjective optimization algorithms (SMOSA, UMOSA, PSA, WDMOSA and PDMOSA) based on SA have been presented. The algorithms are briefly discussed and are compared. The key step of SA is probability calculation, which involves building the annealing schedule. Annealing schedule is discussed briefly. Computational results and suggestions to improve the performance of SA-based multiobjective algorithms are presented. Finally, future research in the area of SA is suggested.