Mask-informed Deep Contrastive Incomplete Multi-view Clustering

TL;DR

This paper introduces Mask-IMvC, a novel multi-view clustering method that effectively handles missing data by using a mask-informed fusion network and contrastive learning to improve clustering accuracy.

Contribution

The paper proposes a new Mask-IMvC approach that integrates incomplete multi-view data using mask-informed fusion and contrastive learning, enhancing clustering performance with missing samples.

Findings

Outperforms state-of-the-art methods on multiple datasets

Effective in both complete and incomplete multi-view scenarios

Improves clustering accuracy with missing data handling

Abstract

Multi-view clustering (MvC) utilizes information from multiple views to uncover the underlying structures of data. Despite significant advancements in MvC, mitigating the impact of missing samples in specific views on the integration of knowledge from different views remains a critical challenge. This paper proposes a novel Mask-informed Deep Contrastive Incomplete Multi-view Clustering (Mask-IMvC) method, which elegantly identifies a view-common representation for clustering. Specifically, we introduce a mask-informed fusion network that aggregates incomplete multi-view information while considering the observation status of samples across various views as a mask, thereby reducing the adverse effects of missing values. Additionally, we design a prior knowledge-assisted contrastive learning loss that boosts the representation capability of the aggregated view-common representation by injecting neighborhood information of samples from different views. Finally, extensive experiments are conducted to demonstrate the superiority of the proposed Mask-IMvC method over state-of-the-art approaches across multiple MvC datasets, both in complete and incomplete scenarios.