`

Timezone: »

 
Poster
Kernel Feature Selection via Conditional Covariance Minimization
Jianbo Chen · Mitchell Stern · Martin J Wainwright · Michael Jordan

Mon Dec 04 06:30 PM -- 10:30 PM (PST) @ Pacific Ballroom #45 #None
in Posters Mon »

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.

Keywords: Nonlinear Dimensionality Reduction and Manifold Learning · Kernel Methods
[ Paper]  [ Poster] 

Author Information

Jianbo Chen (University of California, Berkeley)
Mitchell Stern (UC Berkeley)
Martin J Wainwright (UC Berkeley)
Michael Jordan (UC Berkeley)

More from the Same Authors