Joint Feature Selection and Subspace Learning
Quanquan Gu, Zhenhui Li, Jiawei Han · 2011
Dimensionality reduction is a very important topic in machine learning. It can be generally classi-fied into two categories: feature selection and sub-space learning. In the past decades, many meth-ods have been proposed for dimensionality reduc-tion. However, most of these works study fea-ture selection and subspace learning independently. In this paper, we present a framework for joint feature selection and subspace learning. We re-formulate the subspace learning problem and use L2,1-norm on the projection matrix to achieve row-sparsity, which leads to selecting relevant features and learning transformation simultaneously. We discuss two situations of the proposed framework, and present their optimization algorithms. Experi-ments on benchmark face recognition data sets il-lustrate that the proposed framework outperforms the state of the art methods overwhelmingly. 1