We introduce a framework of feature selection based on dependence maximization between the selected features and the labels of an estimation problem, using the Hilbert-Schmidt Independence Criterion. The key idea is that good features should be highly dependent on the labels. Our approach leads to a greedy procedure for feature selection. We show that a number of existing feature selectors are special cases of this framework. Experiments on both artificial and real-world data show that our feature selector works well in practice.
| Author(s): | Song, L. and Smola, A. and Gretton, A. and Bedo, J. and Borgwardt, K. |
| Journal: | Journal of Machine Learning Research |
| Volume: | 13 |
| Pages: | 1393-1434 |
| Year: | 2012 |
| Month: | May |
| Day: | 0 |
| Bibtex Type: | Article (article) |
| Digital: | 0 |
| Electronic Archiving: | grant_archive |
| Links: | |
BibTex
@article{SongSGBB2012,
title = {Feature Selection via Dependence Maximization},
journal = {Journal of Machine Learning Research},
abstract = {We introduce a framework of feature selection based on dependence maximization between the selected features and the labels of an estimation problem, using the Hilbert-Schmidt Independence Criterion. The key idea is that good features should be highly dependent on the labels. Our approach leads to a greedy procedure for feature selection. We show that
a number of existing feature selectors are special cases of this framework. Experiments on both artificial and real-world data show that our feature selector works well in practice.},
volume = {13},
pages = {1393-1434},
month = may,
year = {2012},
slug = {songsgbb2012},
author = {Song, L. and Smola, A. and Gretton, A. and Bedo, J. and Borgwardt, K.},
month_numeric = {5}
}