不可微合成函数的极小化方法

本文提出了一种极小化不可微合成函数的下降算法。该算法通过内部迭代寻找下降方向，每次内部迭代求解一个二次规划。外部迭代点不精确线搜索求得，算法在有限步内得到近似平稳点，经过适当修正后，算法全局收敛到平衡点。     

一类非精确线性搜索共轭梯度新算法

本文通过对迭代参数的适当选取，给出了一类共轭梯度新算法。在算法的迭代过程中，迭代方向保持下降性，在一般的非精确线性搜索条件下，算法的全局收敛性得到了证明。     

无约束优化的超记忆梯度算法及其收敛特征

研究一种新的无约束优化超记忆梯度算法，算法在每步迭代中充分利用前面迭代点的信息产生下降方向，利用Wolfe线性搜索产生步长，在较弱的条件下证明了算法的全局收敛性。新算法在每步迭代中不需计算和存储矩阵，适于求解大规模优化问题。     

张量方法在信赖域算法中的应用

信赖域算法是求解无约束优化问题的一种有效的算法.对于该算法的子问题,本文将原来目标函数的二次模型扩展成四次张量模型,提出了一个带信赖域约束的四次张量模型优化问题的求解算法.该方法的最大特点是:不仅在张量模型的非稳定点可以得到下降方向及相应的迭代步长,而且在非局部极小值点的稳定点也可以得到下降方向及相应的迭代步长,从而在算法产生的迭代点列中存在一个子列收敛到信赖域子问题的局部极小值点.     

一类非单调曲线搜索方法及其收敛性

本文提出一类求解无约束优化问题的非单调曲线搜索方法, 在较弱条件下证明了其收敛性.该算法有如下特点:(1)采用曲线搜索方法, 在每步迭代时同时确定下降方向和步长;(2)采用非单调搜索技巧, 产生较大的迭代步长, 降低了算法的计算量;(3)利用当前和前面迭代点的信息产生下降方向, 无需计算和存储矩阵, 适于求解大型优化问题.     

一类新的曲线搜索下的多步下降算法

提出一类新的曲线搜索下的多步下降算法,在较弱条件下证明了算法具有全局收敛性和线性收敛速率.算法利用前面多步迭代点的信息和曲线搜索技巧产生新的迭代点,收敛稳定,不用计算和存储矩阵,适于求解大规模优化问题.数值试验表明算法是有效的.     

一类新的具有充分下降条件和强收敛性的共轭梯度法

本文研究了求解无约束优化问题的WYL共轭梯度法.利用修正迭代格式,得到了算法在每步迭代能产生不依赖于搜索条件的充分下降方向.同时,在原算法中关于Wolfe条件中参数去掉的情况下,获得了本文算法是强收敛的.数值实验说明本文算法可以有效求解测试问题.     

一个基于近似下降方向的免梯度算法

本构造一个求解非线性无约束优化问题的免梯度算法，该算法基于传统的模矢法，每次不成功迭代后，充分利用已有迭代点的信息，构造近似下降方向，产生新的迭代点。在较弱条件下，算法是总体收敛的。通过数值实验与传统模矢法相比，计算量明显减少。     

一类新的具有充分下降条件和强收敛性的共轭梯度法（英文）

本文研究了求解无约束优化问题的WYL共轭梯度法.利用修正迭代格式,得到了算法在每步迭代能产生不依赖于搜索条件的充分下降方向.同时,在原算法中关于Wolfe条件中参数去掉的情况下,获得了本文算法是强收敛的.数值实验说明本文算法可以有效求解测试问题.     

等式约束优化非单调信赖域算法

设计了一个新的求解等式约束优化问题的非单调信赖域算法.该算法不需要罚函数也无需滤子.在每次迭代过程中只需求解满足下降条件的拟法向步及切向步.新算法产生的迭代步比滤子方法更易接受,计算量比单调算法小.在一般条件下,算法具有全局收敛性.     

无约束最优化问题中一类新黄族及其性质

本文提出了一类新的用于解决无约束最优化问题的拟牛顿方法,并证明了这样的性质,在 精确线性搜索条件下,每一步该族所有方法所产生的迭代方向和迭代点列仅依赖于参数ρ.该方 法可视为拟牛顿方法中黄族的推广.     

A class of globally convergent conjugate gradient methods

Conjugate gradient methods are very important ones for solving nonlinear optimization problems, especially for large scale problems. However, unlike quasi-Newton methods, conjugate gradient methods were usually analyzed individually. In this paper, we propose a class of conjugate gradient methods, which can be regarded as some kind of convex combination of the Fletcher-Reeves method and the method proposed by Dai et al. To analyze this class of methods, we introduce some unified tools that concern a general method with the scalar βk having the form of φk/φk-1. Consequently, the class of conjugate gradient methods can uniformly be analyzed.     

Multi-step hybrid methods for special second-order differential equations y″(t) = f(t,y(t))

In this paper, multi-step hybrid methods for solving special second-order differential equations y^″(t) = f(t,y(t)) are presented and studied. The new methods inherit the frameworks of RKN methods and linear multi-step methods and include two-step hybrid methods proposed by Coleman (IMA J. Numer. Anal. 23, 197–220, 8) as special cases. The order conditions of the methods were derived by using the SN-series defined on the set SNT of SN-trees. Based on the order conditions, we construct two explicit four-step hybrid methods, which are convergent of order six and seven, respectively. Numerical results show that our new methods are more efficient in comparison with the well-known high quality methods proposed in the scientific literature.     

Generalization of Backward Differentiation Formulas for Parallel Computers

We analyze an extension of backward differentiation formulas, used as boundary value methods, that generates a class of methods with nice stability and convergence properties. These methods are obtained starting from the boundary value GBDFs class, and are in the class of EBDF-type methods. We discuss different ways of using these linear multistep formulas in order to have efficient parallel implementations. Numerical experiments show their effectiveness.     

Iterative algorithms for large partitioned linear systems,with applications to image reconstruction

We present a unifying framework for a wide class of iterative methods in numerical linear algebra. In particular, the class of algorithms contains Kaczmarz's and Richardson's methods for the regularized weighted least squares problem with weighted norm. The convergence theory for this class of algorithms yields as corollaries the usual convergence conditions for Kaczmarz's and Richardson's methods. The algorithms in the class may be characterized as being group-iterative, and incorporate relaxation matrices, as opposed to a single relaxation parameter. We show that some well-known iterative methods of image reconstruction fall into the class of algorithms under consideration, and are thus covered by the convergence theory. We also describe a novel application to truly three-dimensional image reconstruction.     

Explicit symplectic multidimensional exponential fitting modified Runge-Kutta-Nyström methods

This paper is concerned with multidimensional exponential fitting modified Runge-Kutta-Nyström (MEFMRKN) methods for the system of oscillatory second-order differential equations q″(t)+Mq(t)=f(q(t)), where M is a d×d symmetric and positive semi-definite matrix and f(q) is the negative gradient of a potential scalar U(q). We formulate MEFMRKN methods and show clearly the relationship between MEFMRKN methods and multidimensional extended Runge-Kutta-Nyström (ERKN) methods proposed by Wu et al. (Comput. Phys. Comm. 181:1955–1962, 2010). Taking into account the fact that the oscillatory system is a separable Hamiltonian system with Hamiltonian \(H(p,q)=\frac{1}{2}p^{T}p+ \frac{1}{2}q^{T}Mq+U(q)\), we derive the symplecticity conditions for the MEFMRKN methods. Two explicit symplectic MEFMRKN methods are proposed. Numerical experiments accompanied demonstrate that our explicit symplectic MEFMRKN methods are more efficient than some well-known numerical methods appeared in the scientific literature.     

Smooth methods of multipliers for complementarity problems

This paper describes several methods for solving nonlinear complementarity problems. A general duality framework for pairs of monotone operators is developed and then applied to the monotone complementarity problem, obtaining primal, dual, and primal-dual formulations. We derive Bregman-function-based generalized proximal algorithms for each of these formulations, generating three classes of complementarity algorithms. The primal class is well-known. The dual class is new and constitutes a general collection of methods of multipliers, or augmented Lagrangian methods, for complementarity problems. In a special case, it corresponds to a class of variational inequality algorithms proposed by Gabay. By appropriate choice of Bregman function, the augmented Lagrangian subproblem in these methods can be made continuously differentiable. The primal-dual class of methods is entirely new and combines the best theoretical features of the primal and dual methods. Some preliminary computation shows that this class of algorithms is effective at solving many of the standard complementarity test problems. Received February 21, 1997 / Revised version received December 11, 1998? Published online May 12, 1999     

Natural Runge-Kutta and projection methods

Summary Recently the author defined the class of natural Runge-Kutta methods and observed that it includes all the collocation methods. The present paper is devoted to a complete characterization of this class and it is shown that it coincides with the class of the projection methods in some polynomial spaces.This work was supported by the Italian Ministero della Pubblica Istruzione, funds 40%     

A CLASS OF GENERALIZED MULTISPLITTING RELAXATION METHODS FOR LINEAR COMPLEMENTARITY PROBLEMS

In this paper,a class of generalized parallel matrix multisplitting relaxation methods for solving linear complementarity problems on the high-speed multiprocessor systems is set up. This class of methods not only includes all the existing relaxation methods for the linear complementarity problems ,but also yields a lot of novel ones in the sense of multisplittlng. We establish the convergence theories of this class of generalized parallel multisplitting relaxation methods under the condition that the system matrix is an H-metrix with positive diagonal elements.     

一类多值多导数方法的收敛性与稳定性

1 引言 对于多值多导数方法,由于其多值多导的结构特点有利于提高解的精度,以及其包容性大,它包含了当今常用的多种常微数值方法,诸如:线性多步法,单支方法,多步多导方法,多(单)步Runge—Kutta方法,多导Runge-Kutta方法以及混合方法等.因此收敛性与稳定性的研究具有重要的实践意义和广泛的理论指导意义,也正因如此,这方面的研究工作引起了众多数值工作者们的兴趣,近年来,多值多导法求解刚性问题的B—收敛及其非线性稳定性的研究工作巳获得较大进展,其相应成果可参见文献[1—3],在文献[4,5]中笔者则针对Banach空间中一类非刚性问题-K~((p))类问题,分别探讨了多步多导法及单支方法的收敛性