Adaptive Graph Convolutional Subspace Clustering

TL;DR

This paper introduces AGCSC, an adaptive graph convolutional approach that enhances feature extraction and coefficient matrix constraints for improved spectral subspace clustering performance.

Contribution

The paper proposes a novel adaptive graph convolutional method that simultaneously updates feature representations and coefficient constraints for better subspace clustering.

Findings

AGCSC outperforms related methods in experiments.

The adaptive graph convolution improves subspace structure revelation.

AGCSC achieves superior clustering accuracy on benchmark datasets.

Abstract

Spectral-type subspace clustering algorithms have shown excellent performance in many subspace clustering applications. The existing spectral-type subspace clustering algorithms either focus on designing constraints for the reconstruction coefficient matrix or feature extraction methods for finding latent features of original data samples. In this paper, inspired by graph convolutional networks, we use the graph convolution technique to develop a feature extraction method and a coefficient matrix constraint simultaneously. And the graph-convolutional operator is updated iteratively and adaptively in our proposed algorithm. Hence, we call the proposed method adaptive graph convolutional subspace clustering (AGCSC). We claim that by using AGCSC, the aggregated feature representation of original data samples is suitable for subspace clustering, and the coefficient matrix could reveal the subspace structure of the original data set more faithfully. Finally, plenty of subspace clustering experiments prove our conclusions and show that AGCSC outperforms some related methods as well as some deep models.