Convex Subspace Representation Learning from Multi-view Data

Learning from multi-view data is important in many applications. In this paper, we propose a novel convex subspace representation learning method for unsuper-vised multi-view clustering. We first formulate the sub-space learning with multiple views as a joint optimiza-tion problem with a common subspace representation matrix and a group sparsity inducing norm. By exploit-ing the properties of dual norms, we then show a con-vex min-max dual formulation with a sparsity inducing trace norm can be obtained. We develop a proximal bun-dle optimization algorithm to globally solve the min-max optimization problem. Our empirical study shows the proposed subspace representation learning method can effectively facilitate multi-view clustering and in-duce superior clustering results than alternative multi-view clustering methods