序补问题解的存在性

利用耦合不动点的方法得到了混合单调型算子的序补问题解的存在性．同时利用序补问题与隐变分不等式的关系给出了隐变分不等式解的存在性的新条件．     

Multiple positive solutions for one-dimensional p-Laplacian boundary value problems

By means of the Leggett-Williams fixed-point theorem, criteria are developed for the existence of at least three positive solutions to the one-dimensional p-Laplacian boundary value problem, ((y′))′ + g(t)f(t,y) = 0, y(0) - B₀(y′(0)) = 0, y(1) + B₁(y′(1)) = 0, where (v) |v|^p−2v, p > 1.     

Some Results behind Dividend Problems

We consider the basic dividend problem of the compound Poisson model with constant barrierstrategy.Some results concealed behind the dividend problem are made explicit in the present work.Differentmethods and some of which are firstly given in this paper.All these results presented certain direct relationshipbetween some important actuary variables in classical risk theory is also revealed.     

极大单调算子的一个邻近点算法

设E是自反的Banach空间,T∶E→2E是极大单调算子.T-10≠.令x0∈E,yn=(J λnT)-1xn en,xn 1=J-1(αnJxn (1-αn)Jyn),n≥0,λn>0,αn∈[0,1],本文研究了{xn}收敛性.     

一类多线性积分算子的端点有界性

本文对一类相关于非卷积型算子的多线性算子,证明了其在端点情形上的有界性,该算子包括Littlewood-Paley算子和Marcinkiewicz算子．     

不相干算子和强不相干算子的刻画

用算子代数和矩阵论的方法研究不相干量子运算和严格不相干量子运算所对应的kraus算子分解. 首先, 给出有界线性算子是不相干算子的等价条件及不相干算子的具体形式, 并给出用不相干算子刻画不相干量子运算的方法; 其次, 给出强不相干算子的具体形式及用强不相干算子刻画严格不相干量子运算的方法, 并给出严格不相干量子运算与不相干量子运算的关系.     

用遗传算法求解课程表问题

课程表问题是NP完全类问题.近些年来人们尝试着用进化算法求此问题.本文根据大学编排课表的特点设计了一种全新的编码和适应值函数,并应用遗传算法求解.试验说明了该方法的可行性和有效性.     

Generalized indices of operators inB(H)

A generalized dimension is further developed. Here subtraction and addition of two generalized dimensions are defined, so that the operations: ∞ ± n = ∞, ∞ + ∞ = ∞, which used to play an inflexible role, are refined and moreover, ∞ - ∞, which used to be meaningless, is done in sense. Then generalized index for semi-Fred-holm operators is developed to wholeB(H), i.e. all of bounded linear operators in Hilbert spaceH. Theorem 2.2 is proved with an example, which is in contradiction to a known proposition for semi-Fredholm operators in form, practically a refined result of the known proposition. Then, it is proved thatB(H) is the union of countably many disjoint arewise connected sets over all the generalized dimensions ofB(H). Project supported by the National Natural Science Foundation of China     

ON DIRECT METHOD OF SOLUTION FOR A CLASS OF SINGULAR INTEGRAL EQUATIONS

In this article, by introducing characteristic singular integral operator and associate singular integral equations (SIEs), the authors discuss the direct method of solution for a class of singular integral equations with certain analytic inputs. They obtain both the conditions of solvability and the solutions in closed form. It is noteworthy that the method is different from the classical one that is due to Lu.