Deep Incomplete Multi-View Clustering via Hierarchical Imputation and Alignment

2026-01-14

AI Summary

This work addresses the challenge of clustering with partially missing multi-view data by proposing DIMVC-HIA, a novel framework that jointly optimizes latent feature learning, cluster assignment, and view reconstruction through a hierarchical imputation and alignment mechanism. The method innovatively integrates contrastive similarity-based estimation of missing cluster assignments with intra-cluster statistical reconstruction, while introducing energy anchors and a contrastive assignment alignment strategy to ensure cross-view semantic consistency and intra-cluster compactness. DIMVC-HIA unifies view-specific autoencoders, a hierarchical imputation module, and an energy-based semantic alignment component, effectively leveraging the strengths of both deep learning and contrastive learning. Extensive experiments demonstrate that DIMVC-HIA consistently outperforms state-of-the-art methods across various missing rates, achieving superior clustering performance and robustness.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

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
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Incomplete multi-view clustering (IMVC) aims to discover shared cluster structures from multi-view data with partial observations. The core challenges lie in accurately imputing missing views without introducing bias, while maintaining semantic consistency across views and compactness within clusters. To address these challenges, we propose DIMVC-HIA, a novel deep IMVC framework that integrates hierarchical imputation and alignment with four key components: (1) view-specific autoencoders for latent feature extraction, coupled with a view-shared clustering predictor to produce soft cluster assignments; (2) a hierarchical imputation module that first estimates missing cluster assignments based on cross-view contrastive similarity, and then reconstructs missing features using intra-view, intra-cluster statistics; (3) an energy-based semantic alignment module, which promotes intra-cluster compactness by minimizing energy variance around low-energy cluster anchors; and (4) a contrastive assignment alignment module, which enhances cross-view consistency and encourages confident, well-separated cluster predictions. Experiments on benchmarks demonstrate that our framework achieves superior performance under varying levels of missingness.

Problem

• incomplete multi-view clustering
• missing views
• semantic consistency
• cluster compactness
• multi-view data

Innovation

• hierarchical imputation
• semantic alignment
• contrastive learning
• incomplete multi-view clustering
• energy-based modeling