Beyond Independence: Learning Correlated Views for Variational Incomplete Multi-View Clustering

2026-08-16

AI Summary

This study addresses the limitation of conditional view independence in existing variational incomplete multi-view clustering methods by proposing a novel variational framework that explicitly models cross-view correlations. By parameterizing the covariance matrix via normalized Cholesky decomposition, the approach adaptively captures intrinsic data structures through posterior estimation errors under a unified optimization objective. This method overcomes the independence bottleneck with minimal additional parameters and consistently outperforms state-of-the-art baselines across multiple benchmark datasets. These results effectively validate the critical role of explicit cross-view correlation modeling in enhancing clustering performance for incomplete multi-view learning scenarios.

Abstract

Incomplete multi-view clustering (IMVC) aims to uncover shared cluster structures from data with partially observed views. Although recent imputation-free methods based on variational inference demonstrate robustness to missing views, they commonly rely on a conditional independence assumption across views in the posterior aggregation stage, which fails to capture the inherently structured and potentially correlated nature of multi-view data. In this paper, we propose a variational framework that explicitly goes beyond this assumption by introducing a learnable cross-view correlation structure. Specifically, we explicitly model and learn correlations between views by utilizing the covariance structure of posterior estimation errors during aggregation. To facilitate robust and efficient learning, the correlation matrix is parameterized through a normalized Cholesky decomposition, ensuring positive definiteness and enabling the entire model to be trained jointly through a unified variational objective. Extensive experiments on multiple IMVC benchmarks demonstrate that our method consistently outperforms state-of-the-art approaches across diverse missing-view settings while introducing only a negligible number of learnable parameters. These results highlight the effectiveness of adaptive correlation modeling in variational IMVC, demonstrating the need to go beyond the independence assumption in IMVC. The code is available at https://github.com/zmxu196/ACOVA.

Problem

• Incomplete Multi-View Clustering
• Conditional Independence Assumption
• Cross-view Correlation
• Variational Inference

Innovation

• Variational Incomplete Multi-View Clustering
• Cross-view Correlation Structure
• Normalized Cholesky Decomposition
• Posterior Estimation Error Covariance
• Conditional Independence Assumption