Online group-structured dictionary learning

Zoltán Szabó, Barnabás Póczos, A. Lőrincz

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

33 Citations (Scopus)

Abstract

We develop a dictionary learning method which is (i) online, (ii) enables overlapping group structures with (iii) non-convex sparsity-inducing regularization and (iv) handles the partially observable case. Structured sparsity and the related group norms have recently gained widespread attention in group-sparsity regularized problems in the case when the dictionary is assumed to be known and fixed. However, when the dictionary also needs to be learned, the problem is much more difficult. Only a few methods have been proposed to solve this problem, and they can handle two of these four desirable properties at most. To the best of our knowledge, our proposed method is the first one that possesses all of these properties. We investigate several interesting special cases of our framework, such as the online, structured, sparse non-negative matrix factorization, and demonstrate the efficiency of our algorithm with several numerical experiments.

Original languageEnglish
Title of host publicationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Pages2865-2872
Number of pages8
DOIs
Publication statusPublished - 2011
Event2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011 - Colorado Springs, CO, United States
Duration: Jun 20 2011Jun 25 2011

Other

Other2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011
CountryUnited States
CityColorado Springs, CO
Period6/20/116/25/11

Fingerprint

Glossaries
Factorization
Experiments

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Szabó, Z., Póczos, B., & Lőrincz, A. (2011). Online group-structured dictionary learning. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 2865-2872). [5995712] https://doi.org/10.1109/CVPR.2011.5995712

Online group-structured dictionary learning. / Szabó, Zoltán; Póczos, Barnabás; Lőrincz, A.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2011. p. 2865-2872 5995712.

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

Szabó, Z, Póczos, B & Lőrincz, A 2011, Online group-structured dictionary learning. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition., 5995712, pp. 2865-2872, 2011 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2011, Colorado Springs, CO, United States, 6/20/11. https://doi.org/10.1109/CVPR.2011.5995712
Szabó Z, Póczos B, Lőrincz A. Online group-structured dictionary learning. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2011. p. 2865-2872. 5995712 https://doi.org/10.1109/CVPR.2011.5995712
Szabó, Zoltán ; Póczos, Barnabás ; Lőrincz, A. / Online group-structured dictionary learning. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2011. pp. 2865-2872
@inproceedings{7613f08edf7844f29ada434777a7b102,
title = "Online group-structured dictionary learning",
abstract = "We develop a dictionary learning method which is (i) online, (ii) enables overlapping group structures with (iii) non-convex sparsity-inducing regularization and (iv) handles the partially observable case. Structured sparsity and the related group norms have recently gained widespread attention in group-sparsity regularized problems in the case when the dictionary is assumed to be known and fixed. However, when the dictionary also needs to be learned, the problem is much more difficult. Only a few methods have been proposed to solve this problem, and they can handle two of these four desirable properties at most. To the best of our knowledge, our proposed method is the first one that possesses all of these properties. We investigate several interesting special cases of our framework, such as the online, structured, sparse non-negative matrix factorization, and demonstrate the efficiency of our algorithm with several numerical experiments.",
author = "Zolt{\'a}n Szab{\'o} and Barnab{\'a}s P{\'o}czos and A. Lőrincz",
year = "2011",
doi = "10.1109/CVPR.2011.5995712",
language = "English",
isbn = "9781457703942",
pages = "2865--2872",
booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

}

TY - GEN

T1 - Online group-structured dictionary learning

AU - Szabó, Zoltán

AU - Póczos, Barnabás

AU - Lőrincz, A.

PY - 2011

Y1 - 2011

N2 - We develop a dictionary learning method which is (i) online, (ii) enables overlapping group structures with (iii) non-convex sparsity-inducing regularization and (iv) handles the partially observable case. Structured sparsity and the related group norms have recently gained widespread attention in group-sparsity regularized problems in the case when the dictionary is assumed to be known and fixed. However, when the dictionary also needs to be learned, the problem is much more difficult. Only a few methods have been proposed to solve this problem, and they can handle two of these four desirable properties at most. To the best of our knowledge, our proposed method is the first one that possesses all of these properties. We investigate several interesting special cases of our framework, such as the online, structured, sparse non-negative matrix factorization, and demonstrate the efficiency of our algorithm with several numerical experiments.

AB - We develop a dictionary learning method which is (i) online, (ii) enables overlapping group structures with (iii) non-convex sparsity-inducing regularization and (iv) handles the partially observable case. Structured sparsity and the related group norms have recently gained widespread attention in group-sparsity regularized problems in the case when the dictionary is assumed to be known and fixed. However, when the dictionary also needs to be learned, the problem is much more difficult. Only a few methods have been proposed to solve this problem, and they can handle two of these four desirable properties at most. To the best of our knowledge, our proposed method is the first one that possesses all of these properties. We investigate several interesting special cases of our framework, such as the online, structured, sparse non-negative matrix factorization, and demonstrate the efficiency of our algorithm with several numerical experiments.

UR - http://www.scopus.com/inward/record.url?scp=80052890623&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052890623&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2011.5995712

DO - 10.1109/CVPR.2011.5995712

M3 - Conference contribution

AN - SCOPUS:80052890623

SN - 9781457703942

SP - 2865

EP - 2872

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

ER -