Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence

Romain Serizel; Slim Essid; Gael Richard

doi:10.1109/MLSP.2016.7738818

Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence

Romain Serizel, Slim Essid, Gael Richard

Université Paris-Saclay

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

Abstract

Nonnegative matrix factorisation (NMF) with β-divergence is a popular method to decompose real world data. In this paper we propose mini-batch stochastic algorithms to perform NMF efficiently on large data matrices. Besides the stochastic aspect, the mini-batch approach allows exploiting intensive computing devices such as general purpose graphical processing units to decrease the processing time and in some cases outperform coordinate descent approach.

Original language	English
Title of host publication	2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings
Editors	Kostas Diamantaras, Aurelio Uncini, Francesco A. N. Palmieri, Jan Larsen
Publisher	IEEE Computer Society
ISBN (Electronic)	9781509007462
DOIs	https://doi.org/10.1109/MLSP.2016.7738818
Publication status	Published - 8 Nov 2016
Externally published	Yes
Event	26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings - Vietri sul Mare, Salerno, Italy Duration: 13 Sept 2016 → 16 Sept 2016

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume	2016-November
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Conference

Conference	26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings
Country/Territory	Italy
City	Vietri sul Mare, Salerno
Period	13/09/16 → 16/09/16

Keywords

GPGPU
Nonnegative matrix factorisation
multiplicative rules
online learning

Access to Document

10.1109/MLSP.2016.7738818

Cite this

Serizel, R., Essid, S., & Richard, G. (2016). Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence. In K. Diamantaras, A. Uncini, F. A. N. Palmieri, & J. Larsen (Eds.), 2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings Article 7738818 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2016-November). IEEE Computer Society. https://doi.org/10.1109/MLSP.2016.7738818

Serizel, Romain ; Essid, Slim ; Richard, Gael. / Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence. 2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings. editor / Kostas Diamantaras ; Aurelio Uncini ; Francesco A. N. Palmieri ; Jan Larsen. IEEE Computer Society, 2016. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP).

@inproceedings{34954fbdea7345e1aa15709288d99e85,

title = "Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence",

abstract = "Nonnegative matrix factorisation (NMF) with β-divergence is a popular method to decompose real world data. In this paper we propose mini-batch stochastic algorithms to perform NMF efficiently on large data matrices. Besides the stochastic aspect, the mini-batch approach allows exploiting intensive computing devices such as general purpose graphical processing units to decrease the processing time and in some cases outperform coordinate descent approach.",

keywords = "GPGPU, Nonnegative matrix factorisation, multiplicative rules, online learning",

author = "Romain Serizel and Slim Essid and Gael Richard",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings ; Conference date: 13-09-2016 Through 16-09-2016",

year = "2016",

month = nov,

day = "8",

doi = "10.1109/MLSP.2016.7738818",

language = "English",

series = "IEEE International Workshop on Machine Learning for Signal Processing, MLSP",

publisher = "IEEE Computer Society",

editor = "Kostas Diamantaras and Aurelio Uncini and Palmieri, \{Francesco A. N.\} and Jan Larsen",

booktitle = "2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings",

}

Serizel, R, Essid, S & Richard, G 2016, Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence. in K Diamantaras, A Uncini, FAN Palmieri & J Larsen (eds), 2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings., 7738818, IEEE International Workshop on Machine Learning for Signal Processing, MLSP, vol. 2016-November, IEEE Computer Society, 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings, Vietri sul Mare, Salerno, Italy, 13/09/16. https://doi.org/10.1109/MLSP.2016.7738818

Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence. / Serizel, Romain; Essid, Slim; Richard, Gael.
2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings. ed. / Kostas Diamantaras; Aurelio Uncini; Francesco A. N. Palmieri; Jan Larsen. IEEE Computer Society, 2016. 7738818 (IEEE International Workshop on Machine Learning for Signal Processing, MLSP; Vol. 2016-November).

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

TY - GEN

T1 - Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence

AU - Serizel, Romain

AU - Essid, Slim

AU - Richard, Gael

PY - 2016/11/8

Y1 - 2016/11/8

N2 - Nonnegative matrix factorisation (NMF) with β-divergence is a popular method to decompose real world data. In this paper we propose mini-batch stochastic algorithms to perform NMF efficiently on large data matrices. Besides the stochastic aspect, the mini-batch approach allows exploiting intensive computing devices such as general purpose graphical processing units to decrease the processing time and in some cases outperform coordinate descent approach.

AB - Nonnegative matrix factorisation (NMF) with β-divergence is a popular method to decompose real world data. In this paper we propose mini-batch stochastic algorithms to perform NMF efficiently on large data matrices. Besides the stochastic aspect, the mini-batch approach allows exploiting intensive computing devices such as general purpose graphical processing units to decrease the processing time and in some cases outperform coordinate descent approach.

KW - GPGPU

KW - Nonnegative matrix factorisation

KW - multiplicative rules

KW - online learning

U2 - 10.1109/MLSP.2016.7738818

DO - 10.1109/MLSP.2016.7738818

M3 - Conference contribution

AN - SCOPUS:85002368666

T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP

BT - 2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings

A2 - Diamantaras, Kostas

A2 - Uncini, Aurelio

A2 - Palmieri, Francesco A. N.

A2 - Larsen, Jan

PB - IEEE Computer Society

T2 - 26th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings

Y2 - 13 September 2016 through 16 September 2016

ER -

Serizel R, Essid S, Richard G. Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence. In Diamantaras K, Uncini A, Palmieri FAN, Larsen J, editors, 2016 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2016 - Proceedings. IEEE Computer Society. 2016. 7738818. (IEEE International Workshop on Machine Learning for Signal Processing, MLSP). doi: 10.1109/MLSP.2016.7738818

Mini-batch stochastic approaches for accelerated multiplicative updates in nonnegative matrix factorisation with beta-divergence

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this