Deep multi-view sparse subspace clustering

Xiaoliang Tang; Xuan Tang; Wanli Wang; Li Fang; Xian Wei

doi:10.1145/3301326.3301391

Deep multi-view sparse subspace clustering

Xiaoliang Tang
, Xuan Tang
, Wanli Wang
, Li Fang
, Xian Wei

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

22 Scopus citations

Abstract

Most multi-view subspace clustering algorithms construct the affinity matrix with shallow features extracted from each view separately. The integration of multi-view features are left for extended spectral clustering algorithm. The lack of deep feature extraction and interaction across different views prevents the effective exploration of information complementary for multi-view datasets. To address this problem, this paper proposes a novel deep multi-view sparse subspace clustering (DMVSSC) model which consists of convolutional auto-encoders (CAEs) and CCA-based self-expressive module. The proposed model can not only extract deep features of each view data with few parameters but also integrate multi-view features based on CCA. Furthermore, a two-stage joint optimization strategy is proposed for tuning the whole model. Experiments on four benchmark data sets show that our proposed model significantly outperforms the state-of-the-art multi-view subspace clustering algorithms.

Original language	English
Title of host publication	Proceedings of 2018 7th International Conference on Network, Communication and Computing, ICNCC 2018
Publisher	Association for Computing Machinery
Pages	115-119
Number of pages	5
ISBN (Electronic)	9781450365536
DOIs	https://doi.org/10.1145/3301326.3301391
State	Published - 14 Dec 2018
Externally published	Yes
Event	7th International Conference on Network, Communication and Computing, ICNCC 2018 - Taipei, Taiwan, Province of China Duration: 14 Dec 2018 → 16 Dec 2018

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	7th International Conference on Network, Communication and Computing, ICNCC 2018
Country/Territory	Taiwan, Province of China
City	Taipei
Period	14/12/18 → 16/12/18

Keywords

Canonical correlation analysis
Deep convolutional auto-encoder
Multi-view clustering
Sparse subspace clustering

Access to Document

10.1145/3301326.3301391

Cite this

@inproceedings{f2a11b0f01574ae790009389ba3caeae,

title = "Deep multi-view sparse subspace clustering",

abstract = "Most multi-view subspace clustering algorithms construct the affinity matrix with shallow features extracted from each view separately. The integration of multi-view features are left for extended spectral clustering algorithm. The lack of deep feature extraction and interaction across different views prevents the effective exploration of information complementary for multi-view datasets. To address this problem, this paper proposes a novel deep multi-view sparse subspace clustering (DMVSSC) model which consists of convolutional auto-encoders (CAEs) and CCA-based self-expressive module. The proposed model can not only extract deep features of each view data with few parameters but also integrate multi-view features based on CCA. Furthermore, a two-stage joint optimization strategy is proposed for tuning the whole model. Experiments on four benchmark data sets show that our proposed model significantly outperforms the state-of-the-art multi-view subspace clustering algorithms.",

keywords = "Canonical correlation analysis, Deep convolutional auto-encoder, Multi-view clustering, Sparse subspace clustering",

author = "Xiaoliang Tang and Xuan Tang and Wanli Wang and Li Fang and Xian Wei",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computing Machinery.; 7th International Conference on Network, Communication and Computing, ICNCC 2018 ; Conference date: 14-12-2018 Through 16-12-2018",

year = "2018",

month = dec,

day = "14",

doi = "10.1145/3301326.3301391",

language = "英语",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "115--119",

booktitle = "Proceedings of 2018 7th International Conference on Network, Communication and Computing, ICNCC 2018",

address = "美国",

}

Tang, X, Tang, X, Wang, W, Fang, L & Wei, X 2018, Deep multi-view sparse subspace clustering. in Proceedings of 2018 7th International Conference on Network, Communication and Computing, ICNCC 2018. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 115-119, 7th International Conference on Network, Communication and Computing, ICNCC 2018, Taipei, Taiwan, Province of China, 14/12/18. https://doi.org/10.1145/3301326.3301391

Deep multi-view sparse subspace clustering. / Tang, Xiaoliang; Tang, Xuan; Wang, Wanli et al.
Proceedings of 2018 7th International Conference on Network, Communication and Computing, ICNCC 2018. Association for Computing Machinery, 2018. p. 115-119 (ACM International Conference Proceeding Series).

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

TY - GEN

T1 - Deep multi-view sparse subspace clustering

AU - Tang, Xiaoliang

AU - Tang, Xuan

AU - Wang, Wanli

AU - Fang, Li

AU - Wei, Xian

PY - 2018/12/14

Y1 - 2018/12/14

N2 - Most multi-view subspace clustering algorithms construct the affinity matrix with shallow features extracted from each view separately. The integration of multi-view features are left for extended spectral clustering algorithm. The lack of deep feature extraction and interaction across different views prevents the effective exploration of information complementary for multi-view datasets. To address this problem, this paper proposes a novel deep multi-view sparse subspace clustering (DMVSSC) model which consists of convolutional auto-encoders (CAEs) and CCA-based self-expressive module. The proposed model can not only extract deep features of each view data with few parameters but also integrate multi-view features based on CCA. Furthermore, a two-stage joint optimization strategy is proposed for tuning the whole model. Experiments on four benchmark data sets show that our proposed model significantly outperforms the state-of-the-art multi-view subspace clustering algorithms.

AB - Most multi-view subspace clustering algorithms construct the affinity matrix with shallow features extracted from each view separately. The integration of multi-view features are left for extended spectral clustering algorithm. The lack of deep feature extraction and interaction across different views prevents the effective exploration of information complementary for multi-view datasets. To address this problem, this paper proposes a novel deep multi-view sparse subspace clustering (DMVSSC) model which consists of convolutional auto-encoders (CAEs) and CCA-based self-expressive module. The proposed model can not only extract deep features of each view data with few parameters but also integrate multi-view features based on CCA. Furthermore, a two-stage joint optimization strategy is proposed for tuning the whole model. Experiments on four benchmark data sets show that our proposed model significantly outperforms the state-of-the-art multi-view subspace clustering algorithms.

KW - Canonical correlation analysis

KW - Deep convolutional auto-encoder

KW - Multi-view clustering

KW - Sparse subspace clustering

UR - https://www.scopus.com/pages/publications/85063465553

U2 - 10.1145/3301326.3301391

DO - 10.1145/3301326.3301391

M3 - 会议稿件

AN - SCOPUS:85063465553

T3 - ACM International Conference Proceeding Series

SP - 115

EP - 119

BT - Proceedings of 2018 7th International Conference on Network, Communication and Computing, ICNCC 2018

PB - Association for Computing Machinery

T2 - 7th International Conference on Network, Communication and Computing, ICNCC 2018

Y2 - 14 December 2018 through 16 December 2018

ER -

Deep multi-view sparse subspace clustering

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this