Deep Embedded Multi-view Clustering with Collaborative Training

TL;DR

This paper introduces DEMVC, a deep embedded multi-view clustering method that uses collaborative training and a new consistency strategy to improve clustering performance across multiple views.

Contribution

The paper proposes a novel deep embedded multi-view clustering framework with collaborative training and a new cluster center initialization strategy.

Findings

DEMVC outperforms state-of-the-art methods on multiple datasets.

Collaborative training improves the consistency and accuracy of multi-view clustering.

The new cluster center initialization enhances clustering stability and performance.

Abstract

Multi-view clustering has attracted increasing attentions recently by utilizing information from multiple views. However, existing multi-view clustering methods are either with high computation and space complexities, or lack of representation capability. To address these issues, we propose deep embedded multi-view clustering with collaborative training (DEMVC) in this paper. Firstly, the embedded representations of multiple views are learned individually by deep autoencoders. Then, both consensus and complementary of multiple views are taken into account and a novel collaborative training scheme is proposed. Concretely, the feature representations and cluster assignments of all views are learned collaboratively. A new consistency strategy for cluster centers initialization is further developed to improve the multi-view clustering performance with collaborative training. Experimental results on several popular multi-view datasets show that DEMVC achieves significant improvements over state-of-the-art methods.