Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis

Davood Hajinezhad; Mingyi Hong

doi:10.1109/GlobalSIP.2015.7418196

Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis

Davood Hajinezhad, Mingyi Hong

Electrical and Computer Engineering

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

23 Scopus citations

Abstract

In this paper, we propose distributed algorithms to perform sparse principal component analysis (SPCA). The key benefit of the proposed algorithms is their ability to handle distributed data sets. Our algorithms are able to handle a few sparse-promoting regularizers (i.e., the convex norm and the nonconvex log-sum penalty) as well as different forms of data partition (i.e., partition across rows or columns of the data matrix). Our methods are based on a nonconvex ADMM framework, and they are shown to converge to stationary solutions of various nonconvex SPCA formulations. Numerical experiments based on both real and synthetic data sets, conducted on high performance computing (HPC) clusters, demonstrate the effectiveness of our approaches.

Original language	English (US)
Title of host publication	2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	255-259
Number of pages	5
ISBN (Electronic)	9781479975914
DOIs	https://doi.org/10.1109/GlobalSIP.2015.7418196
State	Published - Feb 23 2016
Event	IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 - Orlando, United States Duration: Dec 13 2015 → Dec 16 2015

Publication series

Name	2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

Other

Other	IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
Country/Territory	United States
City	Orlando
Period	12/13/15 → 12/16/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

Distributed Optimization
Non-Convex ADMM
Sparse PCA

Publisher link

10.1109/GlobalSIP.2015.7418196

Find It

Find It at U of M Libraries

Cite this

Hajinezhad, D., & Hong, M. (2016). Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 (pp. 255-259). Article 7418196 (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2015.7418196

Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis. / Hajinezhad, Davood; Hong, Mingyi.
2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 255-259 7418196 (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015).

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

Hajinezhad, D & Hong, M 2016, Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis. in 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015., 7418196, 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015, Institute of Electrical and Electronics Engineers Inc., pp. 255-259, IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015, Orlando, United States, 12/13/15. https://doi.org/10.1109/GlobalSIP.2015.7418196

Hajinezhad D, Hong M. Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 255-259. 7418196. (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015). doi: 10.1109/GlobalSIP.2015.7418196

Hajinezhad, Davood ; Hong, Mingyi. / Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis. 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 255-259 (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015).

@inproceedings{95cb0bc97d8a485885b77428edf616de,

title = "Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis",

abstract = "In this paper, we propose distributed algorithms to perform sparse principal component analysis (SPCA). The key benefit of the proposed algorithms is their ability to handle distributed data sets. Our algorithms are able to handle a few sparse-promoting regularizers (i.e., the convex norm and the nonconvex log-sum penalty) as well as different forms of data partition (i.e., partition across rows or columns of the data matrix). Our methods are based on a nonconvex ADMM framework, and they are shown to converge to stationary solutions of various nonconvex SPCA formulations. Numerical experiments based on both real and synthetic data sets, conducted on high performance computing (HPC) clusters, demonstrate the effectiveness of our approaches.",

keywords = "Distributed Optimization, Non-Convex ADMM, Sparse PCA",

author = "Davood Hajinezhad and Mingyi Hong",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 ; Conference date: 13-12-2015 Through 16-12-2015",

year = "2016",

month = feb,

day = "23",

doi = "10.1109/GlobalSIP.2015.7418196",

language = "English (US)",

series = "2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015",

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

pages = "255--259",

booktitle = "2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015",

}

TY - GEN

T1 - Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis

AU - Hajinezhad, Davood

AU - Hong, Mingyi

PY - 2016/2/23

Y1 - 2016/2/23

N2 - In this paper, we propose distributed algorithms to perform sparse principal component analysis (SPCA). The key benefit of the proposed algorithms is their ability to handle distributed data sets. Our algorithms are able to handle a few sparse-promoting regularizers (i.e., the convex norm and the nonconvex log-sum penalty) as well as different forms of data partition (i.e., partition across rows or columns of the data matrix). Our methods are based on a nonconvex ADMM framework, and they are shown to converge to stationary solutions of various nonconvex SPCA formulations. Numerical experiments based on both real and synthetic data sets, conducted on high performance computing (HPC) clusters, demonstrate the effectiveness of our approaches.

AB - In this paper, we propose distributed algorithms to perform sparse principal component analysis (SPCA). The key benefit of the proposed algorithms is their ability to handle distributed data sets. Our algorithms are able to handle a few sparse-promoting regularizers (i.e., the convex norm and the nonconvex log-sum penalty) as well as different forms of data partition (i.e., partition across rows or columns of the data matrix). Our methods are based on a nonconvex ADMM framework, and they are shown to converge to stationary solutions of various nonconvex SPCA formulations. Numerical experiments based on both real and synthetic data sets, conducted on high performance computing (HPC) clusters, demonstrate the effectiveness of our approaches.

KW - Distributed Optimization

KW - Non-Convex ADMM

KW - Sparse PCA

UR - http://www.scopus.com/inward/record.url?scp=84964703659&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964703659&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2015.7418196

DO - 10.1109/GlobalSIP.2015.7418196

M3 - Conference contribution

AN - SCOPUS:84964703659

T3 - 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

SP - 255

EP - 259

BT - 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

Y2 - 13 December 2015 through 16 December 2015

ER -

Nonconvex alternating direction method of multipliers for distributed sparse principal component analysis

Abstract

Publication series

Other

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this