Weighted Sparse Subspace Representation: A Unified Framework for Subspace Clustering, Constrained Clustering, and Active Learning
This work addresses subspace clustering, constrained clustering, and active learning problems, which are incremental improvements for applications like gene sequencing and image recognition.
The authors tackled the problem of subspace clustering by proposing a novel spectral-based algorithm that represents points as sparse convex combinations of nearby points, extending it to constrained clustering and active learning, and demonstrated its effectiveness through experiments on simulated and real datasets, showing competitive performance with state-of-the-art methods.
Spectral-based subspace clustering methods have proved successful in many challenging applications such as gene sequencing, image recognition, and motion segmentation. In this work, we first propose a novel spectral-based subspace clustering algorithm that seeks to represent each point as a sparse convex combination of a few nearby points. We then extend the algorithm to constrained clustering and active learning settings. Our motivation for developing such a framework stems from the fact that typically either a small amount of labelled data is available in advance; or it is possible to label some points at a cost. The latter scenario is typically encountered in the process of validating a cluster assignment. Extensive experiments on simulated and real data sets show that the proposed approach is effective and competitive with state-of-the-art methods.