Global-Graph Guided and Local-Graph Weighted Contrastive Learning for Unified Clustering on Incomplete and Noise Multi-View Data

2025-12-25

AI Summary

Contrastive learning suffers from sparse and erroneous cross-view sample pairings in incomplete and noisy multi-view data, leading to degraded clustering performance. Method: We propose a novel imputation-free global-local collaborative contrastive clustering framework. It constructs a global view affinity graph to generate reliable cross-view pairs, alleviating pairing sparsity; and introduces a neighbor-driven local graph dynamic weighting mechanism to suppress interference from noisy pairings. The framework jointly optimizes graph neural network embeddings and an adaptive weighted contrastive loss, enabling global-local integrated contrastive learning without data imputation. Contribution/Results: This work establishes the first imputation-free global-local joint contrastive learning paradigm for multi-view clustering. Extensive experiments on multiple incomplete/noisy multi-view benchmarks demonstrate significant improvements over state-of-the-art methods, with average clustering accuracy gains of 3.2–7.8 percentage points, validating the method’s robustness and generalizability.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Recently, contrastive learning (CL) plays an important role in exploring complementary information for multi-view clustering (MVC) and has attracted increasing attention. Nevertheless, real-world multi-view data suffer from data incompleteness or noise, resulting in rare-paired samples or mis-paired samples which significantly challenges the effectiveness of CL-based MVC. That is, rare-paired issue prevents MVC from extracting sufficient multi-view complementary information, and mis-paired issue causes contrastive learning to optimize the model in the wrong direction. To address these issues, we propose a unified CL-based MVC framework for enhancing clustering effectiveness on incomplete and noise multi-view data. First, to overcome the rare-paired issue, we design a global-graph guided contrastive learning, where all view samples construct a global-view affinity graph to form new sample pairs for fully exploring complementary information. Second, to mitigate the mis-paired issue, we propose a local-graph weighted contrastive learning, which leverages local neighbors to generate pair-wise weights to adaptively strength or weaken the pair-wise contrastive learning. Our method is imputation-free and can be integrated into a unified global-local graph-guided contrastive learning framework. Extensive experiments on both incomplete and noise settings of multi-view data demonstrate that our method achieves superior performance compared with state-of-the-art approaches.

Problem

• Enhances clustering on incomplete multi-view data
• Mitigates noise impact in multi-view contrastive learning
• Unifies global and local graph-guided contrastive learning

Innovation

• Global-graph guided contrastive learning for rare-paired samples
• Local-graph weighted contrastive learning for mis-paired samples
• Unified imputation-free framework for incomplete and noisy data