Abstract

In this paper, we introduce and investigate a sparse additive model for subspace clustering problems. Our approach, named SASC (Sparse Additive Subspace Clustering), is essentially a functional exten- sion of the Sparse Subspace Clustering (SSC) of Elhamifar & Vidal [7] to the additive nonparametric setting. To make our model computationally tractable, we express SASC in terms of a finite set of basis functions, and thus the formulated model can be estimated via solving a sequence of grouped Lasso optimization problems. We provide theoretical guar- antees on the subspace recovery performance of our model. Empirical results on synthetic and real data demonstrate the effectiveness of SASC for clustering noisy data points into their original subspaces.