Deep Sparse Subspace Clustering

TL;DR

Deep Sparse Subspace Clustering (DSSC) extends traditional subspace clustering with neural networks, enabling effective clustering of complex data by learning hierarchical nonlinear features, outperforming existing methods.

Contribution

Introduces the first deep learning-based subspace clustering method that combines sparsity and nonlinearity through neural networks.

Findings

DSSC significantly outperforms 12 existing clustering methods.

Neural networks enable subspace assumptions to hold for complex data.

DSSC effectively handles real-world data with nonlinear structures.

Abstract

In this paper, we present a deep extension of Sparse Subspace Clustering, termed Deep Sparse Subspace Clustering (DSSC). Regularized by the unit sphere distribution assumption for the learned deep features, DSSC can infer a new data affinity matrix by simultaneously satisfying the sparsity principle of SSC and the nonlinearity given by neural networks. One of the appealing advantages brought by DSSC is: when original real-world data do not meet the class-specific linear subspace distribution assumption, DSSC can employ neural networks to make the assumption valid with its hierarchical nonlinear transformations. To the best of our knowledge, this is among the first deep learning based subspace clustering methods. Extensive experiments are conducted on four real-world datasets to show the proposed DSSC is significantly superior to 12 existing methods for subspace clustering.