Multi-view Sparse Co-clustering via Proximal Alternating Linearized Minimization

Jiangwen Sun, Jin Lu, Tingyang Xu, Jinbo Bi
Proceedings of the 32nd International Conference on Machine Learning, PMLR 37:757-766, 2015.

Abstract

When multiple views of data are available for a set of subjects, co-clustering aims to identify subject clusters that agree across the different views. We explore the problem of co-clustering when the underlying clusters exist in different subspaces of each view. We propose a proximal alternating linearized minimization algorithm that simultaneously decomposes multiple data matrices into sparse row and columns vectors. This approach is able to group subjects consistently across the views and simultaneously identify the subset of features in each view that are associated with the clusters. The proposed algorithm can globally converge to a critical point of the problem. A simulation study validates that the proposed algorithm can identify the hypothesized clusters and their associated features. Comparison with several latest multi-view co-clustering methods on benchmark datasets demonstrates the superior performance of the proposed approach.

Cite this Paper

BibTeX
@InProceedings{pmlr-v37-sunb15, title = {Multi-view Sparse Co-clustering via Proximal Alternating Linearized Minimization}, author = {Sun, Jiangwen and Lu, Jin and Xu, Tingyang and Bi, Jinbo}, booktitle = {Proceedings of the 32nd International Conference on Machine Learning}, pages = {757--766}, year = {2015}, editor = {Bach, Francis and Blei, David}, volume = {37}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v37/sunb15.pdf}, url = {https://proceedings.mlr.press/v37/sunb15.html}, abstract = {When multiple views of data are available for a set of subjects, co-clustering aims to identify subject clusters that agree across the different views. We explore the problem of co-clustering when the underlying clusters exist in different subspaces of each view. We propose a proximal alternating linearized minimization algorithm that simultaneously decomposes multiple data matrices into sparse row and columns vectors. This approach is able to group subjects consistently across the views and simultaneously identify the subset of features in each view that are associated with the clusters. The proposed algorithm can globally converge to a critical point of the problem. A simulation study validates that the proposed algorithm can identify the hypothesized clusters and their associated features. Comparison with several latest multi-view co-clustering methods on benchmark datasets demonstrates the superior performance of the proposed approach.} }
Endnote
%0 Conference Paper %T Multi-view Sparse Co-clustering via Proximal Alternating Linearized Minimization %A Jiangwen Sun %A Jin Lu %A Tingyang Xu %A Jinbo Bi %B Proceedings of the 32nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Francis Bach %E David Blei %F pmlr-v37-sunb15 %I PMLR %P 757--766 %U https://proceedings.mlr.press/v37/sunb15.html %V 37 %X When multiple views of data are available for a set of subjects, co-clustering aims to identify subject clusters that agree across the different views. We explore the problem of co-clustering when the underlying clusters exist in different subspaces of each view. We propose a proximal alternating linearized minimization algorithm that simultaneously decomposes multiple data matrices into sparse row and columns vectors. This approach is able to group subjects consistently across the views and simultaneously identify the subset of features in each view that are associated with the clusters. The proposed algorithm can globally converge to a critical point of the problem. A simulation study validates that the proposed algorithm can identify the hypothesized clusters and their associated features. Comparison with several latest multi-view co-clustering methods on benchmark datasets demonstrates the superior performance of the proposed approach.
RIS
TY - CPAPER TI - Multi-view Sparse Co-clustering via Proximal Alternating Linearized Minimization AU - Jiangwen Sun AU - Jin Lu AU - Tingyang Xu AU - Jinbo Bi BT - Proceedings of the 32nd International Conference on Machine Learning DA - 2015/06/01 ED - Francis Bach ED - David Blei ID - pmlr-v37-sunb15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 37 SP - 757 EP - 766 L1 - http://proceedings.mlr.press/v37/sunb15.pdf UR - https://proceedings.mlr.press/v37/sunb15.html AB - When multiple views of data are available for a set of subjects, co-clustering aims to identify subject clusters that agree across the different views. We explore the problem of co-clustering when the underlying clusters exist in different subspaces of each view. We propose a proximal alternating linearized minimization algorithm that simultaneously decomposes multiple data matrices into sparse row and columns vectors. This approach is able to group subjects consistently across the views and simultaneously identify the subset of features in each view that are associated with the clusters. The proposed algorithm can globally converge to a critical point of the problem. A simulation study validates that the proposed algorithm can identify the hypothesized clusters and their associated features. Comparison with several latest multi-view co-clustering methods on benchmark datasets demonstrates the superior performance of the proposed approach. ER -
APA
Sun, J., Lu, J., Xu, T. & Bi, J.. (2015). Multi-view Sparse Co-clustering via Proximal Alternating Linearized Minimization. Proceedings of the 32nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 37:757-766 Available from https://proceedings.mlr.press/v37/sunb15.html.

Related Material