"求线性规划问题可行基的一种方法"的再注记

文[1]给出一个求线性规划问题可行基的方法，文[2]指出其判定条件（3）有误，然而所用的反例并不正确。本文给出三个正确的反例；此外，还给出反例表明文[1]的判定条件（2）也不正确的。     

分圆多项式既约因式Φm（x）的快速计算方法

1 引言关于分圆多项式既约因式φm(x)的系数问题,近来《数学通报》连续刊登三篇文章(详见[1]、[2]、[3]进行讨论,为免于如[1]所指出的计算φm(x)时需作大量的多项式除法运算的不足,在文[2]的基础上,本文提出一种速算法,并应用它纠正了文[3]中一个反例φm(x)(m=399)的错误。2 方法     

关于《一类奇异边值问题的正解》的注记

文[4]通过构造反例断言文[1]中定理的必要性证明有误,本文首先指出文[4] 的这个断言不正确,然后对文[4]中定理2.1作了本质性的改进.     

线性两点边值问题的迭加原理与不变嵌入方法(Ⅱ)

在[1]中,对线性两点边值问题,推广了[2一4]中所提出的不变嵌入算法。本文从理论上对[1]中所述方法作进一步讨论並指明[1]中所述的基本假定H在具体计算中的实现方法。最后通过一些实例计算说明我们的方法有较高的效率。为了方便,不加申明地我们将使用[1]中的记号。     

关于线性时变系统的渐近稳定性的一点注记

本文以反例说明文[1]中定理1、2及文[2]中定理1、2不成立. 不失一般性,取向量范数,矩阵范数定义为反例:考虑时变系统     

关于M.R.Taskovi■的不动点定理的注记

本文以反例说明,文[1]中的引理1是错误的.从而,文[1]中的主要定理,即定理1尚不能认为是成立的.     

对《关于几类矩阵的特征值分布》一文的几点意见

文[1]给出了某些对角占优矩阵的特征值分布定理。本文指出文[1]分布定理的几处错误: (1)当矩阵A∈D时,有反例说明定理1的结论2)是不成立的。     

例谈反例在解题中的作用

寻找或构造反例是一种利用特殊化来检验猜想的反面探索.反例的构造非常灵活,文[1]对反例的构建程序作了深入的探索,本文主要谈谈反例在解题中的作用.     

关于有序域的几则短记

Dubois在文[1]中举反例示明,在Artin所解决的Hilbert第17问题中所出现的系数域是个亚基米德有序域这一条件是不能减弱的(见[1]或者[7])在他所举出的反例中,隐约地使用了一些有关序与赋值相互关系的概念与事实,但都没有明显地提出或证明。这篇短文首先是阐明隐含在[1]中的一些概念和事实;其次指出他的例子还可以用来说明     

一个失误的反例

文[1]的作者对于文[2]中的定理2举了一个粗心的反例W_t(x,y)=xy(x-y)(x-ty)．为此,我们不得不与文[1]的作者商榷某些主要问题．     

clever or smart: Strategies for the online target date assignment problem

In this paper, we consider the Online Target Date Assignment Problem (OnlineTDAP) for general downstream problems, where the downstream cost are nonnegative, additive and satisfy the triangle inequality.We analyze algorithm smart, which was introduced by Angelelli et al. [3] and give its exact competitive ratio depending on the number of requests. Since the obtained competitive ratio is at most we answer the question posed in Angelelli et al. [4] if smart has a competitive ratio strictly less than 2.Moreover, we introduce a new algorithm called clever and show that this strategy has a competitive ratio of 3/2. We show that this is asymptotically optimal by proving that no online algorithm can perform better than 3/2−ε.     

NURBS曲线曲面拟合数据点的迭代算法

本文推广了文献[1]的结果,将文献[1]中关于B样条曲线曲面拟合数据点的迭代算法推广至有理形式,给出了无需求解方程组反求控制点及权因子即可得到拟合NURBS曲线曲面的迭代方法．该算法和文献[1]的算法本质上是统一的,而后者恰是前者的一种退化形式．文章还给出了收敛性证明以及一些定性分析．文末的数值实例说明该算法简单实用．     

关于矩阵乘法的一个改进算法的时间复杂度

两个ｎ阶非负整数方阵相乘,常规算法的时间复杂度为Ｏ（ｎ^３）,文献［１］提出一个“运算次数”为Ｏ（ｎ²）的“最佳”算法,文献［２］对此算法做了进一步研究,提出三种改进策略．本文根据算法分析理论,得出改进后的算法的时间复杂度仍不低于Ｏ（ｎ^３ｌｏｇｎ）,因而其阶仍高于常规算法的运算量的阶．     

关于整数向量卷积的一个算法的时间复杂度

众所周知,两个n维整数向量循环卷积的常规算法(即按定义计算)的时间复杂度为O(n~2),现在已有时间复杂度为O(nlog_2n)的快速算法,[1]中提出一个新算法,称其时间复杂度为O(n),因而是最佳的。 本文首先指出[1]的错误原因,再根据算法分析理论得出[1]中算法的时间复杂度不低于O(n~2log_2n),因而比常规算法的运算量还大。     

Computing some classes of Cauchy type singular integrals with Mathematica software

We constructed an algorithm, [SInt], for computing some classes of Cauchy type singular integrals on the unit circle. The design of [SInt] was focused on the possibility of implementing on a computer all the extensive symbolic and numeric calculations present in the algorithm. Furthermore, we show how the factorization algorithm described in Conceição et al. (2010) allowed us to construct and implement the [SIntAFact] algorithm for calculating several interesting singular integrals that cannot be computed by [SInt]. All the above techniques were implemented using the symbolic computation capabilities of the computer algebra system Mathematica. The corresponding source code of [SInt] is made available in this paper. Several examples of nontrivial singular integrals computed with both algorithms are presented.     

Invariant <Emphasis Type="Italic">K</Emphasis>-minimal Sets in the Discrete and Continuous Settings

A sufficient condition for a set \(\Omega \subset L^{1}\left( \left[ 0,1\right] ^{m}\right) \) to be invariant K-minimal with respect to the couple \(\left( L^{1}\left( \left[ 0,1\right] ^{m}\right) ,L^{\infty }\left( \left[ 0,1\right] ^{m}\right) \right) \) is established. Through this condition, different examples of invariant K-minimal sets are constructed. In particular, it is shown that the \(L^{1}\)-closure of the image of the \(L^{\infty }\)-ball of smooth vector fields with support in \(\left( 0,1\right) ^{m}\) under the divergence operator is an invariant K-minimal set. The constructed examples have finite-dimensional analogues in terms of invariant K-minimal sets with respect to the couple \(\left( \ell ^{1},\ell ^{\infty }\right) \) on \(\mathbb {R}^{n}\). These finite-dimensional analogues are interesting in themselves and connected to applications where the element with minimal K-functional is important. We provide a convergent algorithm for computing the element with minimal K-functional in these and other finite-dimensional invariant K-minimal sets.     

The reflexive least squares solutions of the matrix equation A_1X_1B_1+A_2X_2B_2+\cdots +A_lX_lB_l=Cwith a submatrix constraint

In this paper, an efficient algorithm is presented for minimizing $\|A_1X_1B_1 + A_2X_2B_2+\cdots +A_lX_lB_l-C\|$ where $\|\cdot \|$ is the Frobenius norm, $X_i\in R^{n_i \times n_i}(i=1,2,\cdots ,l)$ is a reflexive matrix with a specified central principal submatrix $[x_{ij}]_{r\leq i,j\leq n_i-r}$ . The algorithm produces suitable $[X_1,X_2,\cdots ,X_l]$ such that $\|A_1X_1B_1+A_2X_2B_2+\cdots +A_lX_lB_l-C\|=\min $ within finite iteration steps in the absence of roundoff errors. We show that the algorithm is stable any case. The algorithm requires little storage capacity. Given numerical examples show that the algorithm is efficient.     

On the practical application of the modified romberg algorithm

Experience gained in the use of the modified Romberg algorithm is reported. An alternative form of the algorithm based on a cosine transformation of Euler-MacLaurin's and Euler's second formula is discussed. Some examples where the error estimate incorporated in the algorithm partly fails are given. Finally an ALGOL procedure combining the algorithm discussed in a previous paper of the author [1] and the modification discussed in this paper is described.     

关于《M1-空间的和定理》一文的注记

《数学研究与评论》1988年第一期研究通讯栏目上刊登了论文《M_1-空间的和定理》。我们这篇简短的注记在于说明: (1)上述论文所引进的“强遗传闭包保持集族”的概念等价于“遗传闭包保持集族”的概念。 (2)上述论文中的定理3的结论是不正确的。 关于(1)若正则拓扑空间X的子集族B是X的遗传闭包保持集族,那么由B的元的闭包所组成的集族亦是X的遗传闭包保持集族。     

曲线拐点充分条件证明中的常见错误

文[1]给出了判别曲线拐点的两个充分条件,文[2]给出了一个充分条件,但三个定理的证明都是错误的.同时,文[1]的两个推论也是错误的.本文通过反例分析了其错因,并给出了文[1]中一个拐点充分条件的正确证明.