On optimal low-rank approximation of non-negative matrices

Christian Grussler; Anders Rantzer

doi:10.1109/CDC.2015.7403045

On optimal low-rank approximation of non-negative matrices

Christian Grussler
, Anders Rantzer

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

For low-rank Frobenius-norm approximations of matrices with non-negative entries, it is shown that the Lagrange dual is computable by semi-definite programming. Under certain assumptions the duality gap is zero. Even when the duality gap is non-zero, several new insights are provided.

Original language	English
Title of host publication	54rd IEEE Conference on Decision and Control,CDC 2015
Pages	5278-5283
Number of pages	6
ISBN (Electronic)	9781479978861
DOIs	https://doi.org/10.1109/CDC.2015.7403045
State	Published - 8 Feb 2015
Externally published	Yes
Event	54th IEEE Conference on Decision and Control, CDC 2015 - Osaka, Japan Duration: 15 Dec 2015 → 18 Dec 2015

Publication series

Name	Proceedings of the IEEE Conference on Decision and Control

Conference

Conference	54th IEEE Conference on Decision and Control, CDC 2015
Country/Territory	Japan
City	Osaka
Period	15/12/15 → 18/12/15

Keywords

Computational modeling
Conferences
Context
Convex functions
Image analysis
Programming
Standards

ASJC Scopus subject areas

Control and Systems Engineering
Modelling and Simulation
Control and Optimization

Access to Document

10.1109/CDC.2015.7403045

Cite this

@inproceedings{b5f288b738b241ab864692c82f4b0532,

title = "On optimal low-rank approximation of non-negative matrices",

abstract = "For low-rank Frobenius-norm approximations of matrices with non-negative entries, it is shown that the Lagrange dual is computable by semi-definite programming. Under certain assumptions the duality gap is zero. Even when the duality gap is non-zero, several new insights are provided.",

keywords = "Computational modeling, Conferences, Context, Convex functions, Image analysis, Programming, Standards",

author = "Christian Grussler and Anders Rantzer",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 54th IEEE Conference on Decision and Control, CDC 2015 ; Conference date: 15-12-2015 Through 18-12-2015",

year = "2015",

month = feb,

day = "8",

doi = "10.1109/CDC.2015.7403045",

language = "الإنجليزيّة",

series = "Proceedings of the IEEE Conference on Decision and Control",

pages = "5278--5283",

booktitle = "54rd IEEE Conference on Decision and Control,CDC 2015",

}

TY - GEN

T1 - On optimal low-rank approximation of non-negative matrices

AU - Grussler, Christian

AU - Rantzer, Anders

PY - 2015/2/8

Y1 - 2015/2/8

N2 - For low-rank Frobenius-norm approximations of matrices with non-negative entries, it is shown that the Lagrange dual is computable by semi-definite programming. Under certain assumptions the duality gap is zero. Even when the duality gap is non-zero, several new insights are provided.

AB - For low-rank Frobenius-norm approximations of matrices with non-negative entries, it is shown that the Lagrange dual is computable by semi-definite programming. Under certain assumptions the duality gap is zero. Even when the duality gap is non-zero, several new insights are provided.

KW - Computational modeling

KW - Conferences

KW - Context

KW - Convex functions

KW - Image analysis

KW - Programming

KW - Standards

UR - https://www.scopus.com/pages/publications/84962037361

U2 - 10.1109/CDC.2015.7403045

DO - 10.1109/CDC.2015.7403045

M3 - منشور من مؤتمر

T3 - Proceedings of the IEEE Conference on Decision and Control

SP - 5278

EP - 5283

BT - 54rd IEEE Conference on Decision and Control,CDC 2015

T2 - 54th IEEE Conference on Decision and Control, CDC 2015

Y2 - 15 December 2015 through 18 December 2015

ER -

On optimal low-rank approximation of non-negative matrices

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this