Self-Supervised Discriminative Feature Learning for Deep Multi-View Clustering

TLDR

Multi‑view clustering leverages complementary information from multiple views, yet few methods mitigate the negative impact of views with unclear clustering structures. This work introduces SDMVC, a self‑supervised discriminative feature learning framework for deep multi‑view clustering to overcome that limitation. SDMVC trains deep autoencoders per view, concatenates their embeddings into global features, and uses self‑generated pseudo‑labels to build a unified target distribution that guides all views toward more discriminative, consistent cluster assignments while preserving view diversity. Across a range of multi‑view datasets, SDMVC outperforms 14 baseline methods, including both classic and state‑of‑the‑art approaches. The implementation is publicly available at https://github.com/SubmissionsIn/SDMVC.

Abstract

Multi-view clustering is an important research topic due to its capability to utilize complementary information from multiple views. However, there are few methods to consider the negative impact caused by certain views with unclear clustering structures, resulting in poor multi-view clustering performance. To address this drawback, we propose <u>s</u>elf-supervised discriminative feature learning for <u>d</u>eep <u>m</u>ulti-<u>v</u>iew <u>c</u>lustering (SDMVC). Concretely, deep autoencoders are applied to learn embedded features for each view independently. To leverage the multi-view complementary information, we concatenate all views&#x2019; embedded features to form the global features, which can overcome the negative impact of some views&#x2019; unclear clustering structures. In a self-supervised manner, pseudo-labels are obtained to build a unified target distribution to perform multi-view discriminative feature learning. During this process, global discriminative information can be mined to supervise all views to learn more discriminative features, which in turn are used to update the target distribution. Besides, this unified target distribution can make SDMVC learn consistent cluster assignments, which accomplishes the clustering consistency of multiple views while preserving their features&#x2019; diversity. Experiments on various types of multi-view datasets show that SDMVC outperforms 14 competitors including classic and state-of-the-art methods. The code is available at <uri>https://github.com/SubmissionsIn/SDMVC</uri>.