Deep Incomplete Multi-view Clustering with Distribution Dual-Consistency Recovery Guidance

2025-03-14

AI Summary

In incomplete multi-view clustering, missing views—caused by privacy constraints or device failures—pose significant challenges; existing methods often neglect view heterogeneity, leading to distributional shifts during view recovery, and over-rely on cross-view correlations while ignoring intra-view structural and cluster prototype information. Method: We propose a Distribution Dual-Consistency Recovery Guidance mechanism: (i) modeling sample-level class distributions and enabling cross-view distribution transfer; and (ii) jointly optimizing neighborhood-aware consistency alignment and prototype-based consistency alignment to preserve both intra-view local structure and inter-view global cluster structure. Contribution/Results: Evaluated on multiple benchmark datasets, our method significantly improves both view reconstruction fidelity and clustering accuracy. It is the first to achieve synergistic optimization of high-fidelity, distribution-consistency-driven view reconstruction and end-to-end clustering performance.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Multi-view clustering leverages complementary representations from diverse sources to enhance performance. However, real-world data often suffer incomplete cases due to factors like privacy concerns and device malfunctions. A key challenge is effectively utilizing available instances to recover missing views. Existing methods frequently overlook the heterogeneity among views during recovery, leading to significant distribution discrepancies between recovered and true data. Additionally, many approaches focus on cross-view correlations, neglecting insights from intra-view reliable structure and cross-view clustering structure. To address these issues, we propose BURG, a novel method for incomplete multi-view clustering with distriBution dUal-consistency Recovery Guidance. We treat each sample as a distinct category and perform cross-view distribution transfer to predict the distribution space of missing views. To compensate for the lack of reliable category information, we design a dual-consistency guided recovery strategy that includes intra-view alignment guided by neighbor-aware consistency and cross-view alignment guided by prototypical consistency. Extensive experiments on benchmarks demonstrate the superiority of BURG in the incomplete multi-view scenario.

Problem

• Addresses incomplete multi-view data recovery challenges.
• Proposes dual-consistency guided recovery for missing views.
• Enhances clustering by leveraging intra-view and cross-view structures.

Innovation

• Cross-view distribution transfer for missing views
• Dual-consistency guided recovery strategy
• Intra-view and cross-view alignment mechanisms