On the compression of low rank matrices

被引：218

作者：

Cheng, H

Gimbutas, Z

Martinsson, PG

Rokhlin, V

机构：

[1] MadMax Opt Inc, Hamden, CT 06518 USA

[2] Yale Univ, Dept Math, New Haven, CT 06511 USA

来源：

SIAM JOURNAL ON SCIENTIFIC COMPUTING | 2005年 / 26卷 / 04期

关键词：

matrix factorization; low rank approximation; matrix inversion;

D O I：

10.1137/030602678

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

A procedure is reported for the compression of rank-deficient matrices. A matrix . of rank . is represented in the form . = . circle. circle., where . is a . x . submatrix of ., and . , . are well-conditioned matrices that each contain a . x . identity submatrix. This property enables such compression schemes to be used in certain situations where the singular value decomposition (SVD) cannot be used efficiently. Numerical examples are presented.

引用

页码：1389 / 1404

页数：16

共 10 条

[1] WAVELET-LIKE BASES FOR THE FAST SOLUTION OF 2ND-KIND INTEGRAL-EQUATIONS [J].