AN ALGORITHM FOR A MINIMUM NORM SOLUTION OF A SYSTEM OF LINEAR INEQUALITIES
Authors:
Said Bahi a;
V. P. Sreedharan b
| Affiliations: | a Department of Mathematics and Computer Science, 351 West Center St, Southern Utah University, Cedar City, UT 84720, U.S.A.. |
| b Department of Mathematics, Michigan State University, East Lansing, MI 48824, U.S.A.. |
DOI:
10.1080/0020716021000023079
Publication Frequency:
12 issues per year
Published in:
International Journal of Computer Mathematics,
Volume
80,
Issue
5
May
2003
, pages 639
- 647
Subjects:
Analysis - Mathematics;
Bioinformatics;
Computer Mathematics;
Discrete Mathematics;
Mathematical Finance;
Mathematical Logic;
Mathematical Numerical Analysis;
Systems & Computer Architecture;
Number of References: 10
Formats available:
PDF
(English)
View Article:
View Article (PDF)
Abstract
A common problem encountered in the studies of the least squares problems is that of finding the minimum ℓ2 norm solution of a system of linear equations. In this paper, we consider an algorithm for computing the vector of minimum norm solution of a given system of linear inequalities. We replace the ℓ2 norm by $\ell^p\comma \, 1\lt p \lt\infty$ . Duality theorems and characterizations of the solution are given. The feasibility of the method is proved and some numerical experimentations are included.
|
| Keywords: Minimum Norm Solution; Linear Inequalities; Residual; Smooth Strictly Convex Norm; Least Distance Algorithm |
| view references (10) |

Download Citation

CiteULike
Del.icio.us
BibSonomy
Connotea