Enhanced Latent Multi-View Subspace Clustering
- 17 July 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 34 (12) , 12480-12495
Abstract
Latent multi-view subspace clustering has been demonstrated to have desirable clustering performance. However, the original latent representation method vertically concatenates the data matrices from multiple views into a single matrix along the direction of dimensionality to recover the latent representation matrix, which may result in an incomplete information recovery. To fully recover the latent space representation, we in this paper propose an Enhanced Latent Multi-view Subspace Clustering (ELMSC) method. The ELMSC method involves constructing an augmented data matrix that enhances the representation of multi-view data. Specifically, we stack the data matrices from various views into the block-diagonal locations of the augmented matrix to exploit the complementary information. Meanwhile, the non-block-diagonal entries are composed based on the similarity between different views to capture the consistent information. In addition, we enforce a sparse regularization for the non-diagonal blocks of the augmented self-representation matrix to avoid redundant calculations of consistency information. Finally, a novel iterative algorithm based on the framework of Alternating Direction Method of Multipliers (ADMM) is developed to solve the optimization problem for ELMSC. Particularly, we theoretically analyze the convergence of ELMSC in detail. Extensive experiments on real-world datasets show that our proposed ELMSC is able to achieve higher clustering performance than some state-of-art multi-view clustering methods. Moreover, our experiments show that our method remains effective with randomly chosen parameters, demonstrating ELMSC’s practical potential.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (62201475)
- Natural Science Foundation of Sichuan Province (2024NSFSC1436)
- National Natural Science Foundation of China (U21A20485, 61976175)
This publication has 73 references indexed in Scilit:
- High-order multi-view clustering for generic dataInformation Fusion, 2023
- Frobenius norm-regularized robust graph learning for multi-view subspace clusteringApplied Intelligence, 2022
- Deep multi-view learning methods: A reviewNeurocomputing, 2021
- A novel consensus learning approach to incomplete multi-view clusteringPattern Recognition, 2021
- Multi-view cluster analysis with incomplete data to understand treatment effectsInformation Sciences, 2019
- Global Convergence of ADMM in Nonconvex Nonsmooth OptimizationJournal of Scientific Computing, 2018
- Multi-view low-rank sparse subspace clusteringPattern Recognition, 2018
- Orthogonal self-guided similarity preserving projection for classification and clusteringNeural Networks, 2017
- Principal component analysisWIREs Computational Statistics, 2010
- Introduction to Real Analysis2009