Deep Clustering by Graph Attention Contrastive Learning

Abstract

Contrastive learning shows great potential in deep clustering. It uses constructed pairs to discover the feature distribution that is required for the clustering task. In addition to conventional augmented pairs, recent methods have introduced more methods of creating highly confident pairs, such as nearest neighbors, to provide more semantic prior knowledge. However, existing works only use partial pairwise similarities to construct semantic pairs locally without capturing the entire sample’s relationships from a global perspective. In this paper, we propose a novel clustering framework called graph attention contrastive learning (GACL) to aggregate more semantic information. To this end, GACL is designed to simultaneously perform instance-level and graph-level contrast. Specifically, with its novel graph attention mechanism, our model explores more undiscovered pairs and selectively focuses on informative pairs. To ensure local and global clustering consistency, we jointly use the designed graph-level and instance-level contrastive losses. Experiments on six challenging image benchmarks demonstrate the superiority of our proposed approach over state-of-the-art methods.

Large-Scale Subspace Clustering Based on Purity Kernel Tensor Learning

“…Classic subspace clustering methods typically rely on the self-representation (SE) property of the data, i.e., any data point within the same subspace can be represented as a linear combination of other distinct data points [6]. The goal is to find the minimal number of base points, such that all other points are linear combinations of the base points.…”