B-convergence of a Class of Multistep Multiderivative Methods

Ｂ－ｃｏｎｖｅｒｇｅｎｃｅｏｆａＣｌａｓｓｏｆＭｕｌｔｉｓｔｅｐＭｕｌｔｉｄｅｒｉｖａｔｉｖｅＭｅｔｈｏｄｓＨｕａｎｇＣｈｅｎｇｍｉｎｇ（黄乘明）（ＣｏｍｐｕｔｉｎｇＣｅｎｔｅｒ，ＳｈａｏｙａｎｇＴｅａｃｈｅｒ＇ｓＣｏｌｌｅｇｅ，Ｈｕｎａｎ，４２２０...     

一类具有非单调线搜索的混合共轭梯度算法

提出一类求解无约束最优化问题的混合共轭梯度算法,新算法有机地结合了DY算法和HS算法的优点,并采用非单调线搜索技术在较弱条件下证明了算法的全局收敛性.数值实验表明新算法具有良好的计算效能.     

Parallel Hybrid Methods for a Finite Family of Relatively Nonexpansive Mappings

In this article, we propose two parallel hybrid methods for finding a common fixed point of a finite family of relatively nonexpansive mappings. The strong convergence of the methods is established and their effectiveness are examined by numerical experiments. Thanks to the parallel computation, we can reduce the overall computational effort under widely used assumptions on mappings and spaces.     

A method for rational Chebyshev approximation of rational functions on the unit disk and on the unit interval

A method for construction of CF approximants in some cases of rational approximation of a rational function f on the unit disk and on the unit interval is presented. The inverted square root of the greatest positive eigenvalue and a corresponding eigenvector of an eigenvalue problem defined by the coefficients of f gives the solution.     

A Class of Two-Stage Implicit Hybrid Methods for Ordinary Equations

A k-step, (k+2)th order two-stage implicit hybrid method which has all the advantages of Enright's method but not its principal disadvantages is proposed. A "simple" approach to estimate the local truncation error is developed. Preliminary numerical results indicate that the hybrid method compares favorably with Enright's method.     

一类连分数的有理逼近

设f(n)是非负函数,k,b,s_i,t_i(i=1,2,…)是正常数,研究形如[a_0,a_1,a_2…]=[■]_m~∞=0和[■]_n~∞=1的连分数有理逼近的下界．     

Dogleg路径信赖域方法

本文提出一类新的解无约束最优化问题的信整域方法。这类方法是通过对一般对称矩阵的Bunch-Parlett分解来产生搜索路径。它们既可以解目标函数是二次可微的也可以解目标函数是非二次可微的最优化问题，并且在由算法得到点列的任意聚点上，二次连续可微的目标函数的Hesse阵都是正定或半正定的。我们证明在一些较弱的条件下，算法是整体收敛的；对一致凸函数，是二次收敛的。一些数值结果表明这种新的方法是非常有效的。     

一类Dogleg路径信赖域方法

本文提出一类折线搜索的信赖域方法，用于解无约束最优化问题，这些方法通过对一般对称矩阵的Bunch-Parlett分解来产生搜索路径，我们证明在一些较弱的条件下，算法是整体收敛的，对一致凸函数，是二次收敛的，并且在由算法得到的点列的任意聚点上，连续可微的目标函数的Hesse阵都是正定或半正定的，一些数值结果表明这种新的方法是非常有效的。     

一类共轭梯度算法的全局收敛性

本文证明了一类共轭梯度算法的全局收敛性,其中参数βk满足|βk|≤β,并且αk满足放宽了的强Wolfe线搜索（max｛σ1,σ2｝≤1／2）．     

求解无约束优化问题的一类新的下降算法

本文对求解无约束优化问题提出了一类新的下降算法,并且给出了HS算法与其相结合的两类杂交算法．在Wolfe线搜索下不需给定下降条件,即证明了它们的全局收敛性．数值实验表明新的算法十分有效,尤其是对求解大规模问题而言．     

A RATIONAL SEPCTRAL METHOD FOR SINGULAR DIFFERENTIAL EQUATIONS

An orthogonal system of rational functions is derived from the mapped Laguerre polynomials, which is used for numerical solution of singular differential equations. A model problem is considered. A multiple-step algorithm is developed to implement this method. Numerical results show the efficiency of this new approach.     

A rational approximate solution for generalized pantograph‐delay differential equations

In this paper, a new rational approximation based on a rational interpolation and collocation method is proposed for the solutions of generalized pantograph equations. A comprehensive error analysis is provided. The first part of the error analysis gives an upper bound for the absolute error. The second part is based on residual error procedure that estimates the absolute error. Some numerical examples are given to illustrate the method. The theoretical results support the numerical results. Copyright © 2015 John Wiley & Sons, Ltd.     

具有参数平方收敛的线性插值迭代类

提出了一类具有参数平方收敛的求解非线性方程的线性插值迭代法,方法以Newton法和Steffensen法为其特例,并且给出了该类方法的最佳迭代参数.数值试验表明,选用最佳迭代参数或其近似值的新方法比Newton法和Steffensen方法更有效.     

一类新的插值结点

有界单连通区域Ｇ，其边界θＧ＝Г∈（１，α），α〉０。本计算节以广义Ｆａｂｅｒ多项式φｎ（ｚ）的零点为插值结点的Ｌａｇｒａｎｇｅ插值多项式的逼近性质，得到了它对Ａ（Ｇ↑－）中的函数的一致逼近阶和平均逼近阶的估计，并且得到了它对Ｅ＾ｐ（Ｇ）中函数的平均逼近阶的估计，还指出关于平均逼近阶的估计是不可改进的。     

A unified rational Krylov method for elliptic and parabolic fractional diffusion problems

We present a unified framework to efficiently approximate solutions to fractional diffusion problems of stationary and parabolic type. After discretization, we can take the point of view that the solution is obtained by a matrix-vector product of the form $f^{\mathit{\tau }}(L)\mathbf{b}$, where $L$ is the discretization matrix of the spatial operator, $\mathbf{b}$ a prescribed vector, and $f^{\mathit{\tau }}$ a parametric function, such as a fractional power or the Mittag-Leffler function. In the abstract framework of Stieltjes and complete Bernstein functions, to which the functions we are interested in belong to, we apply a rational Krylov method and prove uniform convergence when using poles based on Zolotarëv's minimal deviation problem. The latter are particularly suited for fractional diffusion as they allow for an efficient query of the map $\mathit{\tau }\mapsto f^{\mathit{\tau }}(L)\mathbf{b}$ and do not degenerate as the fractional parameters approach zero. We also present a variety of both novel and existing pole selection strategies for which we develop a computable error certificate. Our numerical experiments comprise a detailed parameter study of space-time fractional diffusion problems and compare the performance of the poles with the ones predicted by our certificate.     

A Class of Projection Methods for General Variational Inequalities

In this paper, we consider and analyze a new class of projection methods for solving pseudomonotone general variational inequalities using the Wiener-Hopf equations technique. The modified methods converge for pseudomonotone operators. Our proof of convergence is very simple as compared with other methods. The proposed methods include several known methods as special cases.     

有理逼近的一些最新进展(Ⅱ)--倒数逼近的研究综述

倒数逼近作为有理逼近的一种特殊形式，无论在理论上还是在实践中都有着重要的意义，倒数逼近与多项式逼近有本质性的区别，对它的研究有相当大的难度，本文对该领域的一些最新成果和方法作了比较系统的介绍，并提出了一些有待进一步解决的问题。     

无约束优化的一类共轭梯度法(英)

本文给出了一类具有4个参数的共轭梯度法，并且分析了其中两个子类的方法．证明了在步长满足更一般的Wolfe条件时，这两个子类的方法是下降算法．同时还证明了这两个子类算法的全局收敛性．     

部分线性回归模型变窗宽一步局部M-估计

讨论了部分线性回归模型的变窗宽一步局部M-估计.用一步局部M-估计给出未知函数的估计,用平均方法给出参数估计.进一步通过两个引理证明一步M-估计的渐近正态性.所提出的方法继承了局部多项式的优点并且克服了最小二乘法缺乏稳健性的缺点.     

A Class of Potentials for Which Exact Semiclassical Quantization Can Be Achieved

We consider a class of potentials for which the exact semiclassical quantization is achieved by a certain modification of the quantization condition. A list of potentials for which the new quantization condition is exact coincides with the list of potentials for which the spectrum is determined by the factorization method. We construct a one-parameter family of quantization conditions including the supersymmetric WKB condition as a special case. The new condition allows considering the interrelations between different modifications of the leading approximation and their validity ranges and also allows developing new approximate methods for calculating spectra.