MINI-BATCH STOCHASTIC APPROACHES FOR ACCELERATED MULTIPLICATIVE UPDATES IN NONNEGATIVE MATRIX FACTORISATIONWITH BETA-DIVERGENCE

被引：0

作者：

Serizel, Romain ^{[1
]}

Essid, Slim ^{[1
]}

Richard, Gael ^{[1
]}

机构：

[1] Univ Paris Saclay, LTCI, CNRS, Telecom ParisTech, F-75013 Paris, France

来源：

2016 IEEE 26TH INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING (MLSP) | 2016年

关键词：

Nonnegative matrix factorisation; GPGPU; multiplicative rules; online learning; FACTORIZATION; ALGORITHMS;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Nonnegative matrix factorisation (NMF) with beta-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.

引用

页数：6

共 4 条

[1] TOWARDS COMPLEX NONNEGATIVE MATRIX FACTORIZATION WITH THE BETA-DIVERGENCE
Magron, Paul
Virtanen, Tuomas
[J]. 2018 16TH INTERNATIONAL WORKSHOP ON ACOUSTIC SIGNAL ENHANCEMENT (IWAENC), 2018, : 156 - 160
[2] Accelerated Multiplicative Updates and Hierarchical ALS Algorithms for Nonnegative Matrix Factorization
Gillis, Nicolas
Glineur, Francois
[J]. NEURAL COMPUTATION, 2012, 24 (04) : 1085 - 1105
[3] Accelerated stochastic multiplicative update with gradient averaging for nonnegative matrix factorizations
Kasai, Hiroyuki
[J]. 2018 26TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO), 2018, : 2593 - 2597
[4] Spectro-temporal Filtering based on The Beta-divergence for Speech Separation using Nonnegative Matrix Factorization
Fakhry, Mahmoud
[J]. 2021 4TH INTERNATIONAL SEMINAR ON RESEARCH OF INFORMATION TECHNOLOGY AND INTELLIGENT SYSTEMS (ISRITI 2021), 2020,

← 1 →