由FR共轭梯度法控制的两类优化算法的全局收敛性

１引言 考虑无约束优化问题其中ｆ：Ｒｎ→Ｒ是一阶可微函数．求解（１）的非线性共轭梯度法具有如下形式：其中ｇｋ＝　ｆ（ｘｋ）,ａｋ是通过某种线搜索获得的步长,纯量βｋ的选取使得方法（２）—（３）在ｆ（ｘ）是严格凸二次函数且采用精确线搜索时化为线性共轭梯度法［１］．比较常见的βｋ的取法有Ｆｌｅｔｃｈｅｒ－Ｒｅｅｖｅｓ（ＦＲ）公式［２］和Ｐｏｌａｋ－Ｒｉｂｉｅｒｅ－Ｐｏｌｙａｋ（ＰＲＰ）公式［３－４］等．它们分别为其中　　　取欧几里得范数．对于一般非线性函数,ＦＲ方法具有较好的理论收敛性［５－６］,而…     

积分算子的n-逼近数

本文研究了积分算子ＴＫ：Ｌｑ［０，１］→Ｌｑ［０，１］，（ｑ≥１）当核 Ｋ（ｓ， ｔ）是 Ｓｏｂｏｌｅｖ空间 Ｗｐｒ（［０， １］２）中元素时ｎ－逼近数 ａｎ（ＴＫ： Ｌｑ→ Ｌｑ）的估计，并把这个估计应用于退化核方法解第二类线性Ｆｒｅｄｈｏｌｍ方程（Ｉ一ＴＫ）ｘ＝ｙ时，Ｂａｄｈｖａｌｏｖ［５］意义下最佳误差的讨论中，所得到的最佳误差之估计当ｑ＝１时，最优化了［１０］的结论．     

对称Toeplitz系统的快速W变换基预条件子

１．引言考虑下列Ｎ阶线性方程组其中T_N=(t_i,j)　是Ｎ×Ｎ阶实对称正定（ＳＰＤ）Ｔｏｅｐｌｉｔｚ矩阵,即０,１,…,Ｎ－１）且Ｔ_Ｎ的所有特征值为正数．Ｔｏｅｐｌｉｔｚ系统已广泛应用于数字信号处理,时间序列分析（参见［１］）以及微分方程的数值解（参见［２１］等领域．八十年代以前,考虑到Ｔｏｅｐｌｉｔｚ矩阵的特殊性,人们主要用Ｌｅｖｉｎｓｏｎ递推技术及其变形或者分而治之思想直接求解方程组（１．１）,计算复杂性为Ｏ（Ｎ~（２））或Ｏ（ＮｌｏｇＮ~（２））（参见［３］）;比Ｇａｕｓｓ法运算量级Ｏ（Ｎ~（３）…     

θ-方法的非线性渐近稳定性

１引言 数值求解延迟微分方程时,方法的稳定性具有无容置疑的重要性．自１９７５年Ｂａｒｗｅｌｌ引入Ｐ－稳定性与ＧＰ－稳定性概念以来,该领域研究已获许多重要成果（如［７］［８］）．它们大多是基于下面标量线性模型方程：其中λ,μ为复数且满足延迟量τ（＞０）为常数,函数θ（ｔ）连续． 我们首先回忆Ｂａｒｗｅｌｌ［１］的定义． 定义１．１一个数值方法称为是Ｐ－稳定的,如果对任意正整数ｒ用该方法按步长ｈ＝τ／ｒ求解（１．１）时在节点ｔｎ＝ｎｈ的数值解ｙｎ满足 定义１．２一个数值方法称为是ＧＰ－稳定的,如果用该方…     

HILBERT空间中散逸动力系统一般线性方法的散逸稳定性

１．引言 １９９４年,Ｓｔｕａｒｔ与 Ｈｕｍｐｈｒｉｅｓ［４,５］首先考察了用 Ｒｕｎｇｅ－Ｋｕｔｔａ方法求解 Ｒｍ中的散逸动力系统（２．１）－（２．２）时数值解是否继承真解具有的散逸稳定性,并表明代数稳定且不可约的 Ｒｕｎｇｅ－Ｋｕｔｔａ方法是散逸稳定的且有一有界吸引集．１９９６年,本文作者［１］把这一工作推广到了两类特殊的一般线性方法．１９９７年,Ｈｉｌｌ在［３］中证明了Ａ－稳定是单支方法散逸稳定的充要条件,在［２］中又把文［４,５］的工作推广到了 Ｈｉｌｂｅｒｔ空间中的散逸动力系统（２．１）－（…     

关键方程的极小多项式

１引言 本文中Ｒ是指一个ＵＦＤ，ｋ是Ｒ的商域，Ｒ［ｘ］司是以ｘ为未定元的Ｒ上的多项式环．Ｒ上的半无限线性递归序列（ｌｒｓ）与无限线性递归序列（Ｌｒｓ）统记为ＬＲＳ．ＬＲＳ在代数编码、密码学、信号处理中是重要的研究对象，序列的综合问题主要是求出序列ａ的次数最小的特征多项式．在实际应用中，更多地是考察Ｒ上的有限长序列α＝（α_０，α_１，…，α_Ｎ），α（ｘ）＝∑ａ_ｉｘｉ称为α的生成函数．关于求解序列问题的典型描述是解关键方程（ＫｅｙＥｑｕａｔｉｏｎ）：求集合σ＝｛（ｘ）∈Ｒ［ｘ］｜σ（ｘ）ａ（ｘ）≡…     

Fujita型反应扩散方程组整体解的存在性、非存在性与渐近性质

本文研究 Ｆｕｊｉｔａ型反应扩散方程组的初值问题：ｕｔ－△ｕ＝ａ１ｕ~α１－１ｕ＋ｂ１ｖ~β１－１ｖ,ｖｔ－△ｖ＝ａ２ｕ~α２－１ｕ＋ｂ２ｖ~β２－１ｖ,ｕ（Ｘ,０）＝ｕ０（Ｘ）,Ｖ（Ｘ,０）＝Ｖ０（Ｘ）,（Ｘ,ｔ）Ｒ~Ｎ ｘ Ｒ~＋,其中 ａｉ,ｂｉ≥ ０, αｉ,βｉ≥ １（ｉ＝ １,２）,给出了非负整体 Ｌ~ｐ解与古典解存在性与非存在性的一系列充分条件,并讨论了解的渐近性质．本文所用方法和所得结果与已有的工作［１－４］,有很大的不同,不但在某些方面推广了［１－５］,而且从某些方面改进了［１］的结果。     

三维无界区域上耗散Klein-Gordon-Schrodinger方程的指数吸引子

文 ［１］证明了耗散Ｋｌｅｉｎ－Ｇｏｒｄｏｎ－Ｓｃｈｒｏｄｉｎｇｅｒ方程在Ｘ（Ｒ~３）中具有一个最大吸引子,本文在文 ［１］的基础上得到了该方程在Ｘ（Ｒ~３）中拥有一个指数吸引子．     

最佳L_P逼近对P的依赖性

设。本文证明了Ｎ（ｐ）（ｐ≥２）是Ｐ的可微函数，并且Ｎ（ｐ）在区间［１，∞］上满足Ｌｉｐｓｃｈｉｔｚ条件。设是定义在区间［ａ，ｂ］上的ｎ个线性无关的连续函数，记给定ｆ∈Ｌ∞［ａ，ｂ］，令Ｎ（ｐ）＝我们称ｖｐ（１≤Ｐ≤∞）是ｆ的最佳Ｌｐ逼近。     

一个第二类变分不等式的有限元逼近

本短文讨论下述第二类变分不等式（见 ［２, ４］）的有限元逼近及其误差分析：其中是平面凸多边形区域的的边界, 且而 ． 诸如热量控制问题,流体通过半可透性壁的扩散问题以及简化库仑摩擦接触问题的正则化方法等均可归为上述变分不等式（１）（见［２,３］）．在文［２］中给出了上述变分不等式的有限元逼近格式,作出了收敛性分析及误差估计．本文的目的是进一步用数值积分简化上述有限元逼近格式并改进原有的估计误差． 设Ｔｈ是的拟一致三角形部分,Ｖｈ是对应的线性元空间,且使得ｖｈ＝０在上．［２］中用数值积分代替其中 Ｍｉ…     

Geometric Condition Measures and Smoothness Condition Measures for Closed Convex Sets and Linear Regularity of Infinitely Many Closed Convex Sets

In this paper, we study geometric condition measures and smoothness condition measures of closed convex sets, bounded linear regularity, and linear regularity. We show that, under certain conditions, the constant for the linear regularity of infinitely many closed convex sets can be characterized by the geometric condition measure of the intersection or by the smoothness condition measure of the intersection. We study also the bounded linear regularity and present some interesting properties of the general linear regularity problem.The author is grateful to the referees for valuable and constructive suggestions. In particular, she thanks a referee for drawing her attention to Corollary 5.14 of Ref. 3, which inspired her to derive Theorem 4.2 and Corollary 4.2 in the revision of this paper.     

Metric regularity of composition set-valued mappings: Metric setting and coderivative conditions

The paper concerns a new method to obtain a proof of the openness at linear rate/metric regularity of composite set-valued maps on metric spaces by the unification and refinement of several methods developed somehow separately in several works of the authors. In fact, this work is a synthesis and a precise specialization to a general situation of some techniques explored in the last years in the literature. In turn, these techniques are based on several important concepts (like error bounds, lower semicontinuous envelope of a set-valued map, local composition stability of multifunctions) and allow us to obtain two new proofs of a recent result having deep roots in the topic of regularity of mappings. Moreover, we make clear the idea that it is possible to use (co)derivative conditions as tools of proof for openness results in very general situations.     

具线性增长系数的拟线性挠射问题弱解的正则性

讨论了具特殊主部和线性增长系数的n维拟线性抛物挠射问题,利用估计和平均函数方法,证明了弱解在内边界附近的一些正则性质.把这些正则性结果从线性问题推广到这种拟线性问题.     

The convergence of the bilinear and linear immersed finite element solutions to interface problems

This article analyzes the error in both the bilinear and linear immersed finite element (IFE) solutions for second‐order elliptic boundary problems with discontinuous coefficients. The discontinuity in the coefficients is supposed to happen across general curves, but the mesh of the IFE methods can be allowed not to align with the curve of discontinuity. It has been shown that the bilinear and linear IFE solutions converge to the exact solution under the usual assumptions about the meshes and regularity.© 2010 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq 28: 312–330 2012     

带限制线性混合模型中参数估计的小样本性质

本文将带有线性限制下的线性模型理论推广至带有一般线性限制下的线性混合效应模型.同时,本文在没有李(2010)中的正则条件下,构造了估计,考虑了估计的小样本性质.     

Spline collocation methods for linear multi-term fractional differential equations

Some regularity properties of the solution of linear multi-term fractional differential equations are derived. Based on these properties, the numerical solution of such equations by piecewise polynomial collocation methods is discussed. The results obtained in this paper extend the results of Pedas and Tamme (2011) [15] where we have assumed that in the fractional differential equation the order of the highest derivative of the unknown function is an integer. In the present paper, we study the attainable order of convergence of spline collocation methods for solving general linear fractional differential equations using Caputo form of the fractional derivatives and show how the convergence rate depends on the choice of the grid and collocation points. Theoretical results are verified by some numerical examples.     

Spline collocation methods for linear multi-term fractional differential equations

Some regularity properties of the solution of linear multi-term fractional differential equations are derived. Based on these properties, the numerical solution of such equations by piecewise polynomial collocation methods is discussed. The results obtained in this paper extend the results of Pedas and Tamme (2011) [15] where we have assumed that in the fractional differential equation the order of the highest derivative of the unknown function is an integer. In the present paper, we study the attainable order of convergence of spline collocation methods for solving general linear fractional differential equations using Caputo form of the fractional derivatives and show how the convergence rate depends on the choice of the grid and collocation points. Theoretical results are verified by some numerical examples.     

On regularity concepts in variational analysis

The paper offers a new approach that allows unifying various regularity concepts used in variational analysis: the most fundamental local regularity triad (openness at a linear rate – metric regularity – Aubin property), the calmness–subregularity pair, directional regularity, upper Lipschitz continuity etc. The main new element of the approach is the appearance of a new parameter which is a set to which regularity is related (whence the term “relative regularity” used in the paper to name the general property). The main emphasis is put on characterizations of relative regularity and its stability with respect to additive perturbations of the (generally set-valued) mapping. A discussion of a relative extension of strong regularity properties concludes the paper.     

A remark on regularity of powers and products of ideals

We give a simple proof for the fact that the Castelnuovo–Mumford regularity and related invariants of products of powers of ideals are asymptotically linear in the exponents, provided that each ideal is generated by elements of constant degree. We provide examples showing that the asymptotic linearity is false in general. On the other hand, the regularity is always given by the maximum of finitely many linear functions whose coefficients belong to the set of the degrees of generators of the ideals.     

Existence and Regularity for a Class of Nonlinear Hyperbolic Boundary Value Problems

The regularity of the solution of the telegraph system with nonlinear monotone boundary conditions is investigated by two methods. The first one is based on D'Alembert-type representation formulae for the solution. In the second method the telegraph system is reduced to a linear Cauchy problem with a locally Lipschitzian functional perturbation; then regularity results are established by appealing to the theory of linear semigroups.