Hyperbolic Enhanced Representation Learning for Incomplete Multi-view Clustering

2026-04-18

AI Summary

This work addresses the semantic ambiguity in incomplete multi-view clustering caused by missing views and geometric mismatch in Euclidean space by introducing hyperbolic geometry to this task for the first time. The authors propose a structure-aware, doubly constrained hyperbolic contrastive learning framework within the Poincaré ball. Semantic consistency across views is preserved through directional alignment, while hierarchical compactness is enhanced via distance constraints. A hyperbolic prototype head is further designed to correct global structural shifts. This approach effectively disentangles fine-grained semantic associations and achieves substantial improvements over state-of-the-art methods on multiple benchmark datasets, significantly boosting both clustering performance and robustness.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

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
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Incomplete Multi-View Clustering (IMVC) faces the challenge of learning discriminative representations from fragmentary observations while maintaining robustness against missing views. However, prevalent Euclidean-based methods suffer from a geometric mismatch when modeling real-world data with intrinsic hierarchies, leading to semantic blurring where representations drift towards spatially proximal but semantically distinct neighbors. To bridge this gap, we propose HERL, a Hyperbolic Enhanced Representation Learning framework for IMVC. Operating within the Poincaré ball, HERL constructs a structure-aware latent space to enhance representation learning. Specifically, we design a dual-constraint hyperbolic contrastive mechanism optimizing: an angular-based loss to preserve semantic identity via directional alignment, and a distance-based loss to enforce hierarchical compactness. Furthermore, a hyperbolic prototype head is introduced to rectify global structural drift by aligning cross-view hierarchy-aware prototype distributions. Consequently, HERL disentangles fine-grained semantic correlations to sharpen cluster boundaries and imposes geometric constraints to rectify the data recovery process. Extensive experimental results demonstrate that HERL consistently outperforms state-of-the-art approaches.

Problem

• Incomplete Multi-View Clustering
• Representation Learning
• Geometric Mismatch
• Semantic Blurring
• Missing Views

Innovation

• Hyperbolic Geometry
• Incomplete Multi-view Clustering
• Contrastive Learning
• Hierarchical Representation
• Poincaré Ball