full
search.png

Search

close.png

Cancel

arrow
Volume 35, Issue 2
Structured Condition Numbers for the Tikhonov Regularization of Discrete Ill-Posed Problems

Lingsheng Meng & Bing Zheng

DOI: 10.4208/jcm.1608-m2015-0279

J. Comp. Math., 35 (2017), pp. 169-186.

Published online: 2017-04

Preview Purchase PDF 1991 29288

Cited by

google scholar semantic scholar
Export citation
  • Abstract

The possibly most popular regularization method for solving the least squares problem $\mathop{\rm min}\limits_x$$||Ax-b||_2$ with a highly ill-conditioned or rank deficient coefficient matrix $A$ is the Tikhonov regularization method. In this paper we present the explicit expressions of the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when $A$ has linear structures. The structured condition numbers in the special cases of nonlinear structure i.e. Vandermonde and Cauchy matrices are also considered. Some comparisons between structured condition numbers and unstructured condition numbers are made by numerical experiments. In addition, we also derive the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when the coefficient matrix, regularization matrix and right-hand side vector are all perturbed, which generalize the results obtained by Chu et al. [Numer. Linear Algebra Appl., 18 (2011), 87-103].

  • Keywords

Tikhonov regularization, Discrete ill-posed problem, Structured least squares problem, Structured condition number.

  • AMS Subject Headings

65F35, 65F20.

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address

menglsh@nwnu.edu.cn (Lingsheng Meng)

bzheng@lzu.edu.cn (Bing Zheng)

  • BibTex
  • RIS
  • TXT
@Article{JCM-35-169, author = {Meng , Lingsheng and Zheng , Bing}, title = {Structured Condition Numbers for the Tikhonov Regularization of Discrete Ill-Posed Problems}, journal = {Journal of Computational Mathematics}, year = {2017}, volume = {35}, number = {2}, pages = {169--186}, abstract = {

The possibly most popular regularization method for solving the least squares problem $\mathop{\rm min}\limits_x$$||Ax-b||_2$ with a highly ill-conditioned or rank deficient coefficient matrix $A$ is the Tikhonov regularization method. In this paper we present the explicit expressions of the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when $A$ has linear structures. The structured condition numbers in the special cases of nonlinear structure i.e. Vandermonde and Cauchy matrices are also considered. Some comparisons between structured condition numbers and unstructured condition numbers are made by numerical experiments. In addition, we also derive the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when the coefficient matrix, regularization matrix and right-hand side vector are all perturbed, which generalize the results obtained by Chu et al. [Numer. Linear Algebra Appl., 18 (2011), 87-103].

}, issn = {1991-7139}, doi = {https://doi.org/10.4208/jcm.1608-m2015-0279}, url = {http://global-sci.org/intro/article_detail/jcm/9768.html} }
TY - JOUR T1 - Structured Condition Numbers for the Tikhonov Regularization of Discrete Ill-Posed Problems AU - Meng , Lingsheng AU - Zheng , Bing JO - Journal of Computational Mathematics VL - 2 SP - 169 EP - 186 PY - 2017 DA - 2017/04 SN - 35 DO - http://doi.org/10.4208/jcm.1608-m2015-0279 UR - https://global-sci.org/intro/article_detail/jcm/9768.html KW - Tikhonov regularization, Discrete ill-posed problem, Structured least squares problem, Structured condition number. AB -

The possibly most popular regularization method for solving the least squares problem $\mathop{\rm min}\limits_x$$||Ax-b||_2$ with a highly ill-conditioned or rank deficient coefficient matrix $A$ is the Tikhonov regularization method. In this paper we present the explicit expressions of the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when $A$ has linear structures. The structured condition numbers in the special cases of nonlinear structure i.e. Vandermonde and Cauchy matrices are also considered. Some comparisons between structured condition numbers and unstructured condition numbers are made by numerical experiments. In addition, we also derive the normwise, mixed and componentwise condition numbers for the Tikhonov regularization when the coefficient matrix, regularization matrix and right-hand side vector are all perturbed, which generalize the results obtained by Chu et al. [Numer. Linear Algebra Appl., 18 (2011), 87-103].

Meng , Lingsheng and Zheng , Bing. (2017). Structured Condition Numbers for the Tikhonov Regularization of Discrete Ill-Posed Problems. Journal of Computational Mathematics. 35 (2). 169-186. doi:10.4208/jcm.1608-m2015-0279
Copy to clipboard
BibteX RIS TXT
The citation has been copied to your clipboard