Updating the singular value decomposition

Authors:
Philip I. Davies;Matthew I. Smith
Affiliations:
Department of Mathematics, University of Manchester, Manchester, M13 9PL, UK;Department of Mathematics, University of Manchester, Manchester, M13 9PL, UK
Venue:
Journal of Computational and Applied Mathematics
Year:
2004

Citing 6
Cited 3

Topics in matrix analysis

Topics in matrix analysis
Relative perturbation techniques for singular value problems

SIAM Journal on Numerical Analysis
Perturbation Analysis For Two-Sided (or Complete) Orthogonal Decompositions

SIAM Journal on Matrix Analysis and Applications
Matrix computations (3rd ed.)

Matrix computations (3rd ed.)
The symmetric eigenvalue problem

The symmetric eigenvalue problem
Semidefinite Programs: New Search Directions, Smoothing-Type Methods, and Numerical Results

SIAM Journal on Optimization

Fast Spectral Clustering with Random Projection and Sampling

MLDM '09 Proceedings of the 6th International Conference on Machine Learning and Data Mining in Pattern Recognition
Online subspace learning using a misalignment detection for tracking a talking face in video sequence

Neurocomputing
Semi adaptive appearance models for lip tracking

ICIP'09 Proceedings of the 16th IEEE international conference on Image processing

Quantified Score

Hi-index	7.29

Visualization

Abstract

The spectral decomposition of a symmetric matrix A with small off-diagonal and distinct diagonal elements can be approximated using a direct scheme of R. Davies and Modi (Linear Algebra Appl. 77 (1986) 61). In this paper a generalization of this method for computing the singular value decomposition of close-to-diagonal A ∈ Rm×n is presented. When A has repeated or "close" singular values it is possible to apply the direct method to split the problem in two with one part containing the well-separated singular values and one requiring the computation of the "close" singular values.