Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization

Hangjun Che; Jun Wang

doi:10.1109/ICIST.2019.8836758

Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

9 Citations (Scopus)

Abstract

This paper presents a collaborative neurodynamic approach to sparse nonnegative matrix factorization (SNMF). SNMF is formulated as a bilevel optimization problem. In the lower level of the problem, the sparsity of factorized matrix is minimized subject to the factorization error and nonnegative constraints. In the upper level of the problem, the parameter of the inverted Gaussian function is minimized to approximate l₀ norm. A group of neurodynamic models operating at two timescales is employed to solve the reformulated problem. The experimental results show the superiority of the proposed approach.

Original language	English
Title of host publication	9th International Conference on Information Science and Technology - ICIST2019 Final Program
Publisher	IEEE
Pages	114-121
ISBN (Electronic)	978-1-7281-2106-2
DOIs	https://doi.org/10.1109/ICIST.2019.8836758
Publication status	Published - Aug 2019
Event	9th International Conference on Information Science and Technology (ICIST 2019) - Hulunbuir, China Duration: 2 Aug 2019 → 5 Aug 2019

Publication series

Name	International Conference on Information Science and Technology

Conference

Conference	9th International Conference on Information Science and Technology (ICIST 2019)
Abbreviated title	ICIST 2019
Place	China
City	Hulunbuir
Period	2/08/19 → 5/08/19

Research Keywords

Bilevel optimization
Collaborative neurodynamic approach.
Sparse nonnegative matrix factorization

Access Output

10.1109/ICIST.2019.8836758

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{46c43c90c53a4dc9a80f2b765797f2a9,

title = "Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization",

abstract = "This paper presents a collaborative neurodynamic approach to sparse nonnegative matrix factorization (SNMF). SNMF is formulated as a bilevel optimization problem. In the lower level of the problem, the sparsity of factorized matrix is minimized subject to the factorization error and nonnegative constraints. In the upper level of the problem, the parameter of the inverted Gaussian function is minimized to approximate l0 norm. A group of neurodynamic models operating at two timescales is employed to solve the reformulated problem. The experimental results show the superiority of the proposed approach. ",

keywords = "Bilevel optimization, Collaborative neurodynamic approach., Sparse nonnegative matrix factorization",

author = "Hangjun Che and Jun Wang",

year = "2019",

month = aug,

doi = "10.1109/ICIST.2019.8836758",

language = "English",

series = "International Conference on Information Science and Technology ",

publisher = "IEEE",

pages = "114--121",

booktitle = "9th International Conference on Information Science and Technology - ICIST2019 Final Program",

address = "United States",

note = "9th International Conference on Information Science and Technology (ICIST 2019), ICIST 2019 ; Conference date: 02-08-2019 Through 05-08-2019",

}

Che, H & Wang, J 2019, Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization. in 9th International Conference on Information Science and Technology - ICIST2019 Final Program., 8836758, International Conference on Information Science and Technology , IEEE, pp. 114-121, 9th International Conference on Information Science and Technology (ICIST 2019), Hulunbuir, China, 2/08/19. https://doi.org/10.1109/ICIST.2019.8836758

Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization. / Che, Hangjun ; Wang, Jun.
9th International Conference on Information Science and Technology - ICIST2019 Final Program. IEEE, 2019. p. 114-121 8836758 (International Conference on Information Science and Technology ).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization

AU - Che, Hangjun

AU - Wang, Jun

PY - 2019/8

Y1 - 2019/8

N2 - This paper presents a collaborative neurodynamic approach to sparse nonnegative matrix factorization (SNMF). SNMF is formulated as a bilevel optimization problem. In the lower level of the problem, the sparsity of factorized matrix is minimized subject to the factorization error and nonnegative constraints. In the upper level of the problem, the parameter of the inverted Gaussian function is minimized to approximate l0 norm. A group of neurodynamic models operating at two timescales is employed to solve the reformulated problem. The experimental results show the superiority of the proposed approach.

AB - This paper presents a collaborative neurodynamic approach to sparse nonnegative matrix factorization (SNMF). SNMF is formulated as a bilevel optimization problem. In the lower level of the problem, the sparsity of factorized matrix is minimized subject to the factorization error and nonnegative constraints. In the upper level of the problem, the parameter of the inverted Gaussian function is minimized to approximate l0 norm. A group of neurodynamic models operating at two timescales is employed to solve the reformulated problem. The experimental results show the superiority of the proposed approach.

KW - Bilevel optimization

KW - Collaborative neurodynamic approach.

KW - Sparse nonnegative matrix factorization

UR - http://www.scopus.com/inward/record.url?scp=85073218717&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85073218717&origin=recordpage

U2 - 10.1109/ICIST.2019.8836758

DO - 10.1109/ICIST.2019.8836758

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - International Conference on Information Science and Technology

SP - 114

EP - 121

BT - 9th International Conference on Information Science and Technology - ICIST2019 Final Program

PB - IEEE

T2 - 9th International Conference on Information Science and Technology (ICIST 2019)

Y2 - 2 August 2019 through 5 August 2019

ER -

Sparse Nonnegative Matrix Factorization Based on Collaborative Neurodynamic Optimization

Abstract

Publication series

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this