PMSSC: Parallelizable multi-subset based self-expressive model for subspace clustering

TL;DR

This paper introduces PMSSC, a parallelizable subspace clustering method that uses multiple small data subsets for efficient and effective clustering, especially suitable for large datasets.

Contribution

The paper proposes a novel multi-subset self-expressive model (PMS) for subspace clustering that enables parallel computation and improves self-expressiveness by combining multiple coefficient vectors.

Findings

Significant reduction in computational complexity.

Comparable or improved clustering accuracy.

Effective on both synthetic and real-world datasets.

Abstract

Subspace clustering methods which embrace a self-expressive model that represents each data point as a linear combination of other data points in the dataset provide powerful unsupervised learning techniques. However, when dealing with large datasets, representation of each data point by referring to all data points via a dictionary suffers from high computational complexity. To alleviate this issue, we introduce a parallelizable multi-subset based self-expressive model (PMS) which represents each data point by combining multiple subsets, with each consisting of only a small proportion of the samples. The adoption of PMS in subspace clustering (PMSSC) leads to computational advantages because the optimization problems decomposed over each subset are small, and can be solved efficiently in parallel. Furthermore, PMSSC is able to combine multiple self-expressive coefficient vectors obtained from subsets, which contributes to an improvement in self-expressiveness. Extensive experiments on synthetic and real-world datasets show the efficiency and effectiveness of our approach in comparison to other methods.