Imputation-free and Alignment-free: Incomplete Multi-view Clustering Driven by Consensus Semantic Learning

2025-05-16

AI Summary

To address prototype shift and semantic inconsistency caused by view incompleteness in incomplete multi-view clustering (IMVC), this paper proposes a consensus semantic learning paradigm that neither imputes missing views nor enforces explicit alignment. Methodologically, it: (i) jointly learns cross-view shared consensus prototypes from available data to construct a unified semantic space; (ii) incorporates modularity-driven heuristic graph clustering to strengthen intra-view cluster structure; and (iii) designs a contrastive semantic proximity loss to collaboratively optimize multi-view embeddings. By avoiding unreliable imputation and strong consistency assumptions, the approach simultaneously preserves cross-view consensus and view-specific semantics. Extensive experiments on multiple IMVC benchmarks demonstrate significant improvements in clustering accuracy and robustness, yielding more reliable and stable cluster assignments—outperforming current state-of-the-art methods.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

In incomplete multi-view clustering (IMVC), missing data induce prototype shifts within views and semantic inconsistencies across views. A feasible solution is to explore cross-view consistency in paired complete observations, further imputing and aligning the similarity relationships inherently shared across views. Nevertheless, existing methods are constrained by two-tiered limitations: (1) Neither instance- nor cluster-level consistency learning construct a semantic space shared across views to learn consensus semantics. The former enforces cross-view instances alignment, and wrongly regards unpaired observations with semantic consistency as negative pairs; the latter focuses on cross-view cluster counterparts while coarsely handling fine-grained intra-cluster relationships within views. (2) Excessive reliance on consistency results in unreliable imputation and alignment without incorporating view-specific cluster information. Thus, we propose an IMVC framework, imputation- and alignment-free for consensus semantics learning (FreeCSL). To bridge semantic gaps across all observations, we learn consensus prototypes from available data to discover a shared space, where semantically similar observations are pulled closer for consensus semantics learning. To capture semantic relationships within specific views, we design a heuristic graph clustering based on modularity to recover cluster structure with intra-cluster compactness and inter-cluster separation for cluster semantics enhancement. Extensive experiments demonstrate, compared to state-of-the-art competitors, FreeCSL achieves more confident and robust assignments on IMVC task.

Problem

• Addresses semantic inconsistencies in incomplete multi-view clustering
• Eliminates need for imputation and alignment in consensus learning
• Enhances cluster semantics via intra-cluster compactness and separation

Innovation

• Consensus prototypes learning from available data
• Heuristic graph clustering for semantic relationships
• Shared semantic space without imputation or alignment