CaltechAUTHORS
  A Caltech Library Service

Column Subset Selection, Matrix Factorization, and Eigenvalue Optimization

Tropp, Joel A. (2009) Column Subset Selection, Matrix Factorization, and Eigenvalue Optimization. In: Proceedings of the Twentieth Annual ACM-SIAM Symposium on Discrete Algorithms. Association for Computing Machinery , New York, pp. 978-986. ISBN 978-0-898716-80-1. https://resolver.caltech.edu/CaltechAUTHORS:20100921-101535590

[img]
Preview
PDF - Published Version
See Usage Policy.

300Kb
[img] PDF - Submitted Version
See Usage Policy.

255Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20100921-101535590

Abstract

Given a fixed matrix, the problem of column subset selection requests a column submatrix that has favorable spectral properties. Most research from the algorithms and numerical linear algebra communities focuses on a variant called rank-revealing QR, which seeks a well-conditioned collection of columns that spans the (numerical) range of the matrix. The functional analysis literature contains another strand of work on column selection whose algorithmic implications have not been explored. In particular, a celebrated result of Bourgain and Tzafriri demonstrates that each matrix with normalized columns contains a large column submatrix that is exceptionally well conditioned. Unfortunately, standard proofs of this result cannot be regarded as algorithmic. This paper presents a randomized, polynomial-time algorithm that produces the submatrix promised by Bourgain and Tzafriri. The method involves random sampling of columns, followed by a matrix factorization that exposes the well-conditioned subset of columns. This factorization, which is due to Grothendieck, is regarded as a central tool in modern functional analysis. The primary novelty in this work is an algorithm, based on eigenvalue minimization, for constructing the Grothendieck factorization. These ideas also result in an approximation algorithm for the (∞, 1) norm of a matrix, which is generally NP-hard to compute exactly. As an added bonus, this work reveals a surprising connection between matrix factorization and the famous maxcut semidefinite program.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
http://portal.acm.org/citation.cfm?id=1496770.1496876PublisherArticle
https://arxiv.org/abs/0806.4404arXivDiscussion Paper
http://resolver.caltech.edu/CaltechAUTHORS:20111011-161421093Related ItemTechnical Report
ORCID:
AuthorORCID
Tropp, Joel A.0000-0003-1024-1791
Additional Information:© 2009 SIAM. Received 26 June 2008. Revised 2 October 2008. The author thanks Ben Recht for valuable discussions about eigenvalue minimization. Supported in part by ONR award no. N00014-08-1-0883.
Funders:
Funding AgencyGrant Number
Office of Naval Research (ONR)N00014-08-1-0883
Record Number:CaltechAUTHORS:20100921-101535590
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20100921-101535590
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:20069
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:22 Sep 2010 19:54
Last Modified:03 Oct 2019 02:05

Repository Staff Only: item control page