逼近

The first one is to study how the convergence properties of Newton-type methods change with convergent order. The second one is to study what will infects the computation of the inversion of Jacobi matrix.

本项目对其涉及计算复杂性的几个问题进行的研究，关心的问题是： 1）Newton型方法的收敛性随着阶变化的关系。2）寻找是什么影响着Jacobi矩阵逆的逼近。

Inverse eigenvalue problem; Constrained matrix equation problem; Procrustes problem; Matrix norm; Optimal approximation solution

矩阵逆特征值问题；约束矩阵方程问题； Procrustes问题；矩阵范数；最佳逼近解

Keywords constrained matrix equation problem;sub matrix constraint problem; matrix norm; optimal approximation solution;least-squares solution

约束矩阵方程问题；子矩阵约束问题；矩阵范数；最佳逼近解；最小二乘解

In addition, in the solution set of corresponding equation, the expression of the optimalapproximation solution to given matrix is derived.

另外，导出了在线性流形上矩阵方程的解集合中与给定矩阵的最佳逼近解的表达式。

In addition, in the solution set of corresponding equation, the expression of the optimal approximation solution to given matrix is derived.

另外，导出了在矩阵方程的解集合中与给定矩阵的最佳逼近解的表达式。

In addition, the program for the matrix solution was given by using MAT-LAB mathematical software.

同时利用MATLAB数学软件给出相应的计算机求解风x 上一次不定方程组的通用程序，该问题的研究将对函数论和多项式逼近的研究起到一定的作用。

Experimental results demonstrate that the proposed scheme is superior to the scheme based on wavelet modulus maxima by approximating better the edges of images.

实验结果表明，与基于小波模极大值的图像边缘检测方法相比，该文算法有较低的计算复杂度，所提取的边缘更加逼近图像真实边缘。

The ideal rocket represents the maximum performance condition that could be attained if it were not for real-world factors and other approximations that lead to performance reductions in actual solid rocket motors.

理想的火箭，代表最高性能的条件是可以实现的如果不是真实世界的因素和其他的逼近，导致性能降低，在实际固体火箭发动机。

The empirical approximation methods can be applied to study uniform consistency, robustness, strong consistency of the maximum likelihood estimator of statistical estimators.

本文采用的经验逼近方法可应用于研究随机规划统计估计问题的一致相合性、稳健性、极大似然估计的强相合性。

Based on the second PA derivative of mean steady-state system time with respect to a parameter of the service time distribution, we give the second-order stochastic approximation algorithm.

当最优化问题过于复杂，无法使用解析方法分析时，一个可行的方法是在参数的取值范围内用逐次逼近法找到最优点；在使用此类优化算法时，目标函数的样本导数是必需的。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。