Abstract

Low rank matrix completion has been applied successfully in a wide range of machine learning applications, such as collaborative filtering, image inpainting and Microarray data imputation. However, many existing algorithms are not scalable to large-scale problems, as they involve computing singular value decomposition. In this paper, we present an efficient and scalable algo-rithm for matrix completion. The key idea is to extend the well-known orthogonal matching pursuit from the vector case to the matrix case. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish the linear convergence of the proposed algorithm. The fast convergence is achieved due to the proposed construction of matrix bases and the estimation of the weights. We empirically evaluate the proposed algorithm on many real-world large-scale datasets. Results show that our algorithm is much more efficient than state-of-the- Art matrix completion algorithms while achieving similar or better prediction performance.

Original languageEnglish (US)
Title of host publication31st International Conference on Machine Learning, ICML 2014
PublisherInternational Machine Learning Society (IMLS)
Pages1260-1268
Number of pages9
Volume2
ISBN (Print)9781634393973
StatePublished - 2014
Event31st International Conference on Machine Learning, ICML 2014 - Beijing, China
Duration: Jun 21 2014Jun 26 2014

Other

Other31st International Conference on Machine Learning, ICML 2014
CountryChina
CityBeijing
Period6/21/146/26/14

Fingerprint

Collaborative filtering
Singular value decomposition
Microarrays
Learning systems

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Software

Cite this

Wang, Z., Lai, M. J., Lu, Z., Fan, W., Davulcu, H., & Ye, J. (2014). Rank-one matrix pursuit for matrix completion. In 31st International Conference on Machine Learning, ICML 2014 (Vol. 2, pp. 1260-1268). International Machine Learning Society (IMLS).

Rank-one matrix pursuit for matrix completion. / Wang, Zheng; Lai, Ming Jun; Lu, Zhaosong; Fan, Wei; Davulcu, Hasan; Ye, Jieping.

31st International Conference on Machine Learning, ICML 2014. Vol. 2 International Machine Learning Society (IMLS), 2014. p. 1260-1268.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, Z, Lai, MJ, Lu, Z, Fan, W, Davulcu, H & Ye, J 2014, Rank-one matrix pursuit for matrix completion. in 31st International Conference on Machine Learning, ICML 2014. vol. 2, International Machine Learning Society (IMLS), pp. 1260-1268, 31st International Conference on Machine Learning, ICML 2014, Beijing, China, 6/21/14.
Wang Z, Lai MJ, Lu Z, Fan W, Davulcu H, Ye J. Rank-one matrix pursuit for matrix completion. In 31st International Conference on Machine Learning, ICML 2014. Vol. 2. International Machine Learning Society (IMLS). 2014. p. 1260-1268
Wang, Zheng ; Lai, Ming Jun ; Lu, Zhaosong ; Fan, Wei ; Davulcu, Hasan ; Ye, Jieping. / Rank-one matrix pursuit for matrix completion. 31st International Conference on Machine Learning, ICML 2014. Vol. 2 International Machine Learning Society (IMLS), 2014. pp. 1260-1268
@inproceedings{91574e43d561403ba2324b7d37ffe8a2,
title = "Rank-one matrix pursuit for matrix completion",
abstract = "Low rank matrix completion has been applied successfully in a wide range of machine learning applications, such as collaborative filtering, image inpainting and Microarray data imputation. However, many existing algorithms are not scalable to large-scale problems, as they involve computing singular value decomposition. In this paper, we present an efficient and scalable algo-rithm for matrix completion. The key idea is to extend the well-known orthogonal matching pursuit from the vector case to the matrix case. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish the linear convergence of the proposed algorithm. The fast convergence is achieved due to the proposed construction of matrix bases and the estimation of the weights. We empirically evaluate the proposed algorithm on many real-world large-scale datasets. Results show that our algorithm is much more efficient than state-of-the- Art matrix completion algorithms while achieving similar or better prediction performance.",
author = "Zheng Wang and Lai, {Ming Jun} and Zhaosong Lu and Wei Fan and Hasan Davulcu and Jieping Ye",
year = "2014",
language = "English (US)",
isbn = "9781634393973",
volume = "2",
pages = "1260--1268",
booktitle = "31st International Conference on Machine Learning, ICML 2014",
publisher = "International Machine Learning Society (IMLS)",

}

TY - GEN

T1 - Rank-one matrix pursuit for matrix completion

AU - Wang, Zheng

AU - Lai, Ming Jun

AU - Lu, Zhaosong

AU - Fan, Wei

AU - Davulcu, Hasan

AU - Ye, Jieping

PY - 2014

Y1 - 2014

N2 - Low rank matrix completion has been applied successfully in a wide range of machine learning applications, such as collaborative filtering, image inpainting and Microarray data imputation. However, many existing algorithms are not scalable to large-scale problems, as they involve computing singular value decomposition. In this paper, we present an efficient and scalable algo-rithm for matrix completion. The key idea is to extend the well-known orthogonal matching pursuit from the vector case to the matrix case. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish the linear convergence of the proposed algorithm. The fast convergence is achieved due to the proposed construction of matrix bases and the estimation of the weights. We empirically evaluate the proposed algorithm on many real-world large-scale datasets. Results show that our algorithm is much more efficient than state-of-the- Art matrix completion algorithms while achieving similar or better prediction performance.

AB - Low rank matrix completion has been applied successfully in a wide range of machine learning applications, such as collaborative filtering, image inpainting and Microarray data imputation. However, many existing algorithms are not scalable to large-scale problems, as they involve computing singular value decomposition. In this paper, we present an efficient and scalable algo-rithm for matrix completion. The key idea is to extend the well-known orthogonal matching pursuit from the vector case to the matrix case. In each iteration, we pursue a rank-one matrix basis generated by the top singular vector pair of the current approximation residual and update the weights for all rank-one matrices obtained up to the current iteration. We further propose a novel weight updating rule to reduce the time and storage complexity, making the proposed algorithm scalable to large matrices. We establish the linear convergence of the proposed algorithm. The fast convergence is achieved due to the proposed construction of matrix bases and the estimation of the weights. We empirically evaluate the proposed algorithm on many real-world large-scale datasets. Results show that our algorithm is much more efficient than state-of-the- Art matrix completion algorithms while achieving similar or better prediction performance.

UR - http://www.scopus.com/inward/record.url?scp=84919883227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84919883227&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84919883227

SN - 9781634393973

VL - 2

SP - 1260

EP - 1268

BT - 31st International Conference on Machine Learning, ICML 2014

PB - International Machine Learning Society (IMLS)

ER -