1) low-rank decomposition
低秩分解
1.
A low-rank decomposition technique is adopted to transform the standard semidefinite programming into an equivalent nonlinear programming problem.
首先采用低秩分解技术将一般的半定规划问题转化为与其等价的非线性规划问题;然后利用多目标优化中的占优概念,来建立一个有效的筛子,使目标函数和不可行性达到最优,建立了半定规划的筛选法;最后给出了算法的收敛性分析。
2.
Methods A low-rank decomposition technique is adopted to transform the general semidefinite programming into an equivalent nonlinear programming problem.
方法采用低秩分解技术将一般的半定规划问题转化为与其等价的非线性规划问题,利用基于方向分解的筛选算法,通过对搜索方向进行切线步和垂直步的分解来分别寻求最优解方向和不可行性改善的方向,构造了半定规划问题的筛选算法。
2) full rank decomposition
满秩分解
1.
Based on the two simple alternative practical methods of the full rank decomposition and Gaussian elimintion,this paper was primarily concerning with the algorithms for some generalized invers
在两种可供选择的满秩分解方法和Gauss消元法的基础上,主要研究了某些广义逆的计算。
2.
And all of these approaches are derived from the full rank decomposition technique for the incidence matrix.
这些方法都是基于网关联矩阵的满秩分解。
3.
Based on matrix s elementary row operation remaining its column vector s linear relationship and Hermite standard form of matrix,the paper gives a simple method for solving full rank decomposition of matrix only through elementary row operation.
利用矩阵初等行变换不改变矩阵列向量组线性关系的性质,以及矩阵的Hermite标准形,给出了一种只通过初等行变换可求得矩阵满秩分解的简单方法。
3) full rank factorization
满秩分解
1.
The full rank factorization and Moore-Penrose inverse for generalized row(column) unitary symmetric matrix
广义行(列)酉对称矩阵的满秩分解及其Moore-Penrose逆
2.
The concept of row (column) transposed matrix and row (column) symmetric matrix is given,their basic property is studied,and the formula for full rank factorization and orthogonal diagonal factorization of row (column) symmetric matrix are presented,which can reduce dramatically the amount of calculation and save the CPU time and memory without loss of any numerical precision.
提出了行(列)转置矩阵与行(列)对称矩阵的概念,研究了其性质,给出了行(列)对称矩阵的满秩分解和正交对角分解公式,极大地减少了行(列)对称矩阵的满秩分解和正交对角分解的计算量与存储量,且没有降低数值精度。
3.
In addition,the formulas of the full rank factorization,rank factorization and generalized inverse of row (column) antisymmetric matrix are given,which make calculation easier and accurate.
利用分块矩阵理论获得了许多新的结果,给出了行(列)反对称矩阵的满秩分解、秩分解和广义逆的公式及快速算法。
4) rank one decomposition
秩1分解
1.
By using the rank one decomposition method and Kronecker multiplication method the proof of r(A * B)≤r(A)r(B) is presented.
讨论了矩阵Hadamard乘积的一些性质,分别用秩1分解法和Kronecker乘积法给出了r(A*B)≤r(A)r(B)的证明。
6) full-rank decomposition
满秩分解
1.
The necessary and sufficient conditions for the consistency of the equations with an anti-symmetric condition on solutions are derived using the full-rank decomposition,generalized inverse and generalized singular-value decomposition.
分别用满秩分解、广义逆和广义奇异值分解导出了具有反对称解的充分必要条
补充资料：下秩
1.谓官职低。
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条