Distributed dyadic cyclic descent for non-negative matrix factorization

M. O. Ulfarsson; V. Solo; J. Sigurdsson; J. R. Sveinsson

doi:10.1109/ICASSP.2016.7472489

Distributed dyadic cyclic descent for non-negative matrix factorization

M. O. Ulfarsson, V. Solo, J. Sigurdsson, J. R. Sveinsson

Faculty of Electrical and Computer Engineering

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

2 Citations (Scopus)

Abstract

Non-negative matrix factorization (NMF) has found use in fields such as remote sensing and computer vision where the signals of interest are usually non-negative. Data dimensions in these applications can be huge and traditional algorithms break down due to unachievable memory demands. One is then compelled to consider distributed algorithms. In this paper, we develop for the first time a distributed version of NMF using the alternating direction method of multipliers (ADMM) algorithm and dyadic cyclic descent. The algorithm is compared to well established variants of NMF using simulated data, and is also evaluated using real remote sensing hyperspectral data.

Original language	English
Title of host publication	2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	4303-4307
Number of pages	5
ISBN (Electronic)	9781479999880
DOIs	https://doi.org/10.1109/ICASSP.2016.7472489
Publication status	Published - 18 May 2016
Event	41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Shanghai, China Duration: 20 Mar 2016 → 25 Mar 2016

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2016-May
ISSN (Print)	1520-6149

Conference

Conference	41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016
Country/Territory	China
City	Shanghai
Period	20/03/16 → 25/03/16

Bibliographical note

Publisher Copyright:
© 2016 IEEE.

Other keywords

alternating direction method of multipliers
distributed signal processing
dyadic cyclic descent
Non-negative matrix factorization
optimization

Access to Document

10.1109/ICASSP.2016.7472489

Cite this

Ulfarsson, M. O., Solo, V., Sigurdsson, J., & Sveinsson, J. R. (2016). Distributed dyadic cyclic descent for non-negative matrix factorization. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings (pp. 4303-4307). Article 7472489 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2016-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2016.7472489

Ulfarsson, M. O. ; Solo, V. ; Sigurdsson, J. et al. / Distributed dyadic cyclic descent for non-negative matrix factorization. 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 4303-4307 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{aac03189decc4e0c9a18e825a7387699,

title = "Distributed dyadic cyclic descent for non-negative matrix factorization",

abstract = "Non-negative matrix factorization (NMF) has found use in fields such as remote sensing and computer vision where the signals of interest are usually non-negative. Data dimensions in these applications can be huge and traditional algorithms break down due to unachievable memory demands. One is then compelled to consider distributed algorithms. In this paper, we develop for the first time a distributed version of NMF using the alternating direction method of multipliers (ADMM) algorithm and dyadic cyclic descent. The algorithm is compared to well established variants of NMF using simulated data, and is also evaluated using real remote sensing hyperspectral data.",

keywords = "alternating direction method of multipliers, distributed signal processing, dyadic cyclic descent, Non-negative matrix factorization, optimization",

author = "Ulfarsson, {M. O.} and V. Solo and J. Sigurdsson and Sveinsson, {J. R.}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 ; Conference date: 20-03-2016 Through 25-03-2016",

year = "2016",

month = may,

day = "18",

doi = "10.1109/ICASSP.2016.7472489",

language = "English",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "4303--4307",

booktitle = "2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings",

address = "United States",

}

Ulfarsson, MO, Solo, V, Sigurdsson, J & Sveinsson, JR 2016, Distributed dyadic cyclic descent for non-negative matrix factorization. in 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings., 7472489, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2016-May, Institute of Electrical and Electronics Engineers Inc., pp. 4303-4307, 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016, Shanghai, China, 20/03/16. https://doi.org/10.1109/ICASSP.2016.7472489

Distributed dyadic cyclic descent for non-negative matrix factorization. / Ulfarsson, M. O.; Solo, V.; Sigurdsson, J. et al.
2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2016. p. 4303-4307 7472489 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2016-May).

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

TY - GEN

T1 - Distributed dyadic cyclic descent for non-negative matrix factorization

AU - Ulfarsson, M. O.

AU - Solo, V.

AU - Sigurdsson, J.

AU - Sveinsson, J. R.

PY - 2016/5/18

Y1 - 2016/5/18

N2 - Non-negative matrix factorization (NMF) has found use in fields such as remote sensing and computer vision where the signals of interest are usually non-negative. Data dimensions in these applications can be huge and traditional algorithms break down due to unachievable memory demands. One is then compelled to consider distributed algorithms. In this paper, we develop for the first time a distributed version of NMF using the alternating direction method of multipliers (ADMM) algorithm and dyadic cyclic descent. The algorithm is compared to well established variants of NMF using simulated data, and is also evaluated using real remote sensing hyperspectral data.

AB - Non-negative matrix factorization (NMF) has found use in fields such as remote sensing and computer vision where the signals of interest are usually non-negative. Data dimensions in these applications can be huge and traditional algorithms break down due to unachievable memory demands. One is then compelled to consider distributed algorithms. In this paper, we develop for the first time a distributed version of NMF using the alternating direction method of multipliers (ADMM) algorithm and dyadic cyclic descent. The algorithm is compared to well established variants of NMF using simulated data, and is also evaluated using real remote sensing hyperspectral data.

KW - alternating direction method of multipliers

KW - distributed signal processing

KW - dyadic cyclic descent

KW - Non-negative matrix factorization

KW - optimization

UR - http://www.scopus.com/inward/record.url?scp=84973338725&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2016.7472489

DO - 10.1109/ICASSP.2016.7472489

M3 - Conference contribution

AN - SCOPUS:84973338725

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 4303

EP - 4307

BT - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 41st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016

Y2 - 20 March 2016 through 25 March 2016

ER -

Ulfarsson MO, Solo V, Sigurdsson J, Sveinsson JR. Distributed dyadic cyclic descent for non-negative matrix factorization. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2016 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2016. p. 4303-4307. 7472489. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2016.7472489

Distributed dyadic cyclic descent for non-negative matrix factorization

Abstract

Publication series

Conference

Bibliographical note

Other keywords

Access to Document

Fingerprint

Cite this