Global Graph Propagation with Hierarchical Information Transfer for Incomplete Contrastive Multi-view Clustering

2025-02-26

AI Summary

Incomplete multi-view clustering faces practical challenges due to ubiquitous view missingness, yet existing methods struggle to effectively model implicit information in missing views and commonly adopt a suboptimal two-stage paradigm—separating representation learning from clustering. Method: We propose an end-to-end joint optimization framework: (i) view-specific graph convolutional networks to capture local structures; (ii) consensus-representation-driven global graph propagation and hierarchical information transfer to bridge missing views; and (iii) weight-sharing pseudo-classifiers with contrastive clustering loss to unify representation learning, missing-view modeling, and clustering. Contribution/Results: This is the first work to integrate global graph propagation and hierarchical information transfer into incomplete multi-view clustering. Extensive experiments on multiple benchmark datasets demonstrate significant improvements over state-of-the-art methods in both clustering accuracy and robustness.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Representation Learning for Vision
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Incomplete multi-view clustering has become one of the important research problems due to the extensive missing multi-view data in the real world. Although the existing methods have made great progress, there are still some problems: 1) most methods cannot effectively mine the information hidden in the missing data; 2) most methods typically divide representation learning and clustering into two separate stages, but this may affect the clustering performance as the clustering results directly depend on the learned representation. To address these problems, we propose a novel incomplete multi-view clustering method with hierarchical information transfer. Firstly, we design the view-specific Graph Convolutional Networks (GCN) to obtain the representation encoding the graph structure, which is then fused into the consensus representation. Secondly, considering that one layer of GCN transfers one-order neighbor node information, the global graph propagation with the consensus representation is proposed to handle the missing data and learn deep representation. Finally, we design a weight-sharing pseudo-classifier with contrastive learning to obtain an end-to-end framework that combines view-specific representation learning, global graph propagation with hierarchical information transfer, and contrastive clustering for joint optimization. Extensive experiments conducted on several commonly-used datasets demonstrate the effectiveness and superiority of our method in comparison with other state-of-the-art approaches. The code is available at https://github.com/KelvinXuu/GHICMC.

Problem

• Addresses incomplete multi-view clustering challenges
• Integrates representation learning and clustering stages
• Utilizes hierarchical information transfer for missing data

Innovation

• Graph Convolutional Networks enhance representation
• Global graph propagation handles missing data
• Contrastive learning optimizes clustering end-to-end