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

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