Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition

Niranjay Ravindran; Nicholas D. Sidiropoulos; Shaden Smith; George Karypis

doi:10.1109/ACSSC.2014.7094512

Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition

Niranjay Ravindran, Nicholas D. Sidiropoulos, Shaden Smith, George Karypis

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

13 Scopus citations

Abstract

Low-rank tensor decomposition has many applications in signal processing and machine learning, and is becoming increasingly important for analyzing big data. A significant challenge is the computation of intermediate products which can be much larger than the final result of the computation, or even the original tensor. We propose a scheme that allows memory-efficient in-place updates of intermediate matrices. Motivated by recent advances in big tensor decomposition from multiple compressed replicas, we also consider the related problem of memory-efficient tensor compression. The resulting algorithms can be parallelized, and can exploit but do not require sparsity.

Original language	English (US)
Title of host publication	Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers
Editors	Michael B. Matthews
Publisher	IEEE Computer Society
Pages	581-585
Number of pages	5
ISBN (Electronic)	9781479982974
DOIs	https://doi.org/10.1109/ACSSC.2014.7094512
State	Published - Apr 24 2015
Event	48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015 - Pacific Grove, United States Duration: Nov 2 2014 → Nov 5 2014

Publication series

Name	Conference Record - Asilomar Conference on Signals, Systems and Computers
Volume	2015-April
ISSN (Print)	1058-6393

Other

Other	48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015
Country/Territory	United States
City	Pacific Grove
Period	11/2/14 → 11/5/14

Publisher link

10.1109/ACSSC.2014.7094512

Find It

Find It at U of M Libraries

Cite this

Ravindran, N., Sidiropoulos, N. D., Smith, S., & Karypis, G. (2015). Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition. In M. B. Matthews (Ed.), Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers (pp. 581-585). [7094512] (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2015-April). IEEE Computer Society. https://doi.org/10.1109/ACSSC.2014.7094512

Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition. / Ravindran, Niranjay; Sidiropoulos, Nicholas D.; Smith, Shaden et al.
Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers. ed. / Michael B. Matthews. IEEE Computer Society, 2015. p. 581-585 7094512 (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2015-April).

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

Ravindran, N, Sidiropoulos, ND, Smith, S & Karypis, G 2015, Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition. in MB Matthews (ed.), Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers., 7094512, Conference Record - Asilomar Conference on Signals, Systems and Computers, vol. 2015-April, IEEE Computer Society, pp. 581-585, 48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015, Pacific Grove, United States, 11/2/14. https://doi.org/10.1109/ACSSC.2014.7094512

Ravindran N, Sidiropoulos ND, Smith S, Karypis G. Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition. In Matthews MB, editor, Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society. 2015. p. 581-585. 7094512. (Conference Record - Asilomar Conference on Signals, Systems and Computers). doi: 10.1109/ACSSC.2014.7094512

Ravindran, Niranjay ; Sidiropoulos, Nicholas D. ; Smith, Shaden et al. / Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition. Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers. editor / Michael B. Matthews. IEEE Computer Society, 2015. pp. 581-585 (Conference Record - Asilomar Conference on Signals, Systems and Computers).

@inproceedings{3dac7c1b9a8944a590cc32de60560ca0,

title = "Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition",

abstract = "Low-rank tensor decomposition has many applications in signal processing and machine learning, and is becoming increasingly important for analyzing big data. A significant challenge is the computation of intermediate products which can be much larger than the final result of the computation, or even the original tensor. We propose a scheme that allows memory-efficient in-place updates of intermediate matrices. Motivated by recent advances in big tensor decomposition from multiple compressed replicas, we also consider the related problem of memory-efficient tensor compression. The resulting algorithms can be parallelized, and can exploit but do not require sparsity.",

author = "Niranjay Ravindran and Sidiropoulos, {Nicholas D.} and Shaden Smith and George Karypis",

year = "2015",

month = apr,

day = "24",

doi = "10.1109/ACSSC.2014.7094512",

language = "English (US)",

series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",

publisher = "IEEE Computer Society",

pages = "581--585",

editor = "Matthews, {Michael B.}",

booktitle = "Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers",

note = "48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015 ; Conference date: 02-11-2014 Through 05-11-2014",

}

TY - GEN

T1 - Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition

AU - Ravindran, Niranjay

AU - Sidiropoulos, Nicholas D.

AU - Smith, Shaden

AU - Karypis, George

PY - 2015/4/24

Y1 - 2015/4/24

N2 - Low-rank tensor decomposition has many applications in signal processing and machine learning, and is becoming increasingly important for analyzing big data. A significant challenge is the computation of intermediate products which can be much larger than the final result of the computation, or even the original tensor. We propose a scheme that allows memory-efficient in-place updates of intermediate matrices. Motivated by recent advances in big tensor decomposition from multiple compressed replicas, we also consider the related problem of memory-efficient tensor compression. The resulting algorithms can be parallelized, and can exploit but do not require sparsity.

AB - Low-rank tensor decomposition has many applications in signal processing and machine learning, and is becoming increasingly important for analyzing big data. A significant challenge is the computation of intermediate products which can be much larger than the final result of the computation, or even the original tensor. We propose a scheme that allows memory-efficient in-place updates of intermediate matrices. Motivated by recent advances in big tensor decomposition from multiple compressed replicas, we also consider the related problem of memory-efficient tensor compression. The resulting algorithms can be parallelized, and can exploit but do not require sparsity.

UR - http://www.scopus.com/inward/record.url?scp=84940476299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940476299&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2014.7094512

DO - 10.1109/ACSSC.2014.7094512

M3 - Conference contribution

AN - SCOPUS:84940476299

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 581

EP - 585

BT - Conference Record of the 48th Asilomar Conference on Signals, Systems and Computers

A2 - Matthews, Michael B.

PB - IEEE Computer Society

T2 - 48th Asilomar Conference on Signals, Systems and Computers, ACSSC 2015

Y2 - 2 November 2014 through 5 November 2014

ER -

Memory-efficient parallel computation of tensor and matrix products for big tensor decomposition

Abstract

Publication series

Other

Publisher link

Find It

Other files and links

Fingerprint

Cite this