Contrastive Continual Multiview Clustering With Filtered Structural Fusion
- 4 December 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 36 (8) , 14829-14842
Abstract
Multiview clustering thrives in applications where views are collected in advance by extracting consistent and complementary information among views. However, it overlooks scenarios where data views are collected sequentially, i.e., real-time data. Due to privacy issues or memory burden, previous views are not available with time in these situations. Some methods are proposed to handle it but are trapped in a stability-plasticity dilemma. In specific, these methods undergo a catastrophic forgetting of prior knowledge when a new view is attained. Such a catastrophic forgetting problem (CFP) would cause the consistent and complementary information hard to get and affect the clustering performance. To tackle this, we propose a novel method termed contrastive continual multiview clustering with filtered structural fusion (CCMVC-FSF). Precisely, considering that data correlations play a vital role in clustering and prior knowledge ought to guide the clustering process of a new view, we develop a data buffer to store filtered structural information and utilize it to guide the generation of a robust partition matrix via contrastive learning. Additionally, to address the high complexity involved in acquiring and storing structural information, we propose a sampling strategy called clustering then sample. Furthermore, we theoretically connect CCMVC-FSF with semisupervised learning and knowledge distillation. Extensive experiments exhibit the excellence of the proposed method. Our code is publicly available at https://github.com/wanxinhang/CCMVC-FSF/.Keywords
All Related Versions
Funding Information
- National Science and Technology Innovation 2030 Major Project (2022ZD0209103)
- National Natural Science Foundation of China (62071157, 62276271, 62306324, 62325604, 62376039, 62376279, 62421002, 62476281)
- science and technology innovation Program of Hunan Province (2024RC3128)
- National University of Defense Technology Research Foundation (ZK24-30)
- Independent Innovation Science Fund Project of the National University of Defense Technology
This publication has 67 references indexed in Scilit:
- Structured anchor-inferred graph learning for universal incomplete multi-view clusteringWorld Wide Web, 2022
- Incremental multi-view spectral clustering with sparse and connected graph learningNeural Networks, 2021
- Deep embedded multi-view clustering with collaborative trainingInformation Sciences, 2021
- Understanding the Behaviour of Contrastive Loss2021
- Partially View-aligned Representation Learning with Noise-robust Contrastive Loss2021
- Multi-view information fusion in mammograms: A comprehensive overviewInformation Fusion, 2019
- AE2-Nets: Autoencoder in Autoencoder Networks2019
- Incremental multi-view spectral clusteringKnowledge-Based Systems, 2019
- Multi-view low-rank matrix factorization using multiple manifold regularizationNeurocomputing, 2019
- A generalized power iteration method for solving quadratic problem on the Stiefel manifoldScience China Information Sciences, 2017