Kernel Feature Selection via Conditional Covariance Minimization

We propose a framework for feature selection that employs kernel-based measures of independence to find a subset of covariates that is maximally predictive of the response. Building on past work in kernel dimension reduction, we formulate our approach as a constrained optimization problem involving the trace of the conditional covariance operator, and additionally provide some consistency results. We then demonstrate on a variety of synthetic and real data sets that our method compares favorably with other state-of-the-art algorithms.