A scaling algorithm to equilibrate both rows and columns norms in matrices

ePubs

The open archive for STFC research publications

Persistent URL	http://purl.org/net/epubs/work/29557
Record Status	Checked
Record Id	29557
Title	A scaling algorithm to equilibrate both rows and columns norms in matrices
Contributors	D Ruiz
Abstract	We present an iterative procedure which asymptotically scales the infinity norm of both rows and columns in a matrix to 1. This scaling strategy exhibits some optimality properties and additionally preserves symmetry. The algorithm also shows fast linear convergence with an asymptotic rate of 1=2. We discuss possible extensions of such an algorithm when considering the one-norm or the two norm of the rows and columns of the given matrix, and give the proof of its convergence when the matrix pattern satisfies some common properties.
Organisation	CCLRC
Keywords
Funding Information
Related Research Object(s):
Licence Information:
Language	English (EN)

Type	Details	URI(s)	Local file(s)	Year
Report	RAL Technical Reports RAL-TR-2001-034. 2001.		raltr-2001034.pdf	2001