Kernel Feature Selection via Conditional Covariance Minimization
Machine Learning
2018-10-23 v2 Artificial Intelligence
Machine Learning
Methodology
Abstract
We propose a method 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 show how to perform feature selection via a constrained optimization problem involving the trace of the conditional covariance operator. We prove various consistency results for this procedure, and also demonstrate that our method compares favorably with other state-of-the-art algorithms on a variety of synthetic and real data sets.
Cite
@article{arxiv.1707.01164,
title = {Kernel Feature Selection via Conditional Covariance Minimization},
author = {Jianbo Chen and Mitchell Stern and Martin J. Wainwright and Michael I. Jordan},
journal= {arXiv preprint arXiv:1707.01164},
year = {2018}
}
Comments
The first two authors contributed equally