Feature selection using e-values

Subhabrata Majumdar, Snigdhansu Chatterjee
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:14753-14773, 2022.

Abstract

In the context of supervised learning, we introduce the concept of e-value. An e-value is a scalar quantity that represents the proximity of the sampling distribution of parameter estimates in a model trained on a subset of features to that of the model trained on all features (i.e. the full model). Under general conditions, a rank ordering of e-values separates models that contain all essential features from those that do not. For a p-dimensional feature space, this requires fitting only the full model and evaluating p+1 models, as opposed to the traditional requirement of fitting and evaluating 2^p models. The above e-values framework is applicable to a wide range of parametric models. We use data depths and a fast resampling-based algorithm to implement a feature selection procedure, providing consistency results. Through experiments across several model settings and synthetic and real datasets, we establish that the e-values can be a promising general alternative to existing model-specific methods of feature selection.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-majumdar22a, title = {Feature selection using e-values}, author = {Majumdar, Subhabrata and Chatterjee, Snigdhansu}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {14753--14773}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/majumdar22a/majumdar22a.pdf}, url = {https://proceedings.mlr.press/v162/majumdar22a.html}, abstract = {In the context of supervised learning, we introduce the concept of e-value. An e-value is a scalar quantity that represents the proximity of the sampling distribution of parameter estimates in a model trained on a subset of features to that of the model trained on all features (i.e. the full model). Under general conditions, a rank ordering of e-values separates models that contain all essential features from those that do not. For a p-dimensional feature space, this requires fitting only the full model and evaluating p+1 models, as opposed to the traditional requirement of fitting and evaluating 2^p models. The above e-values framework is applicable to a wide range of parametric models. We use data depths and a fast resampling-based algorithm to implement a feature selection procedure, providing consistency results. Through experiments across several model settings and synthetic and real datasets, we establish that the e-values can be a promising general alternative to existing model-specific methods of feature selection.} }
Endnote
%0 Conference Paper %T Feature selection using e-values %A Subhabrata Majumdar %A Snigdhansu Chatterjee %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-majumdar22a %I PMLR %P 14753--14773 %U https://proceedings.mlr.press/v162/majumdar22a.html %V 162 %X In the context of supervised learning, we introduce the concept of e-value. An e-value is a scalar quantity that represents the proximity of the sampling distribution of parameter estimates in a model trained on a subset of features to that of the model trained on all features (i.e. the full model). Under general conditions, a rank ordering of e-values separates models that contain all essential features from those that do not. For a p-dimensional feature space, this requires fitting only the full model and evaluating p+1 models, as opposed to the traditional requirement of fitting and evaluating 2^p models. The above e-values framework is applicable to a wide range of parametric models. We use data depths and a fast resampling-based algorithm to implement a feature selection procedure, providing consistency results. Through experiments across several model settings and synthetic and real datasets, we establish that the e-values can be a promising general alternative to existing model-specific methods of feature selection.
APA
Majumdar, S. & Chatterjee, S.. (2022). Feature selection using e-values. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:14753-14773 Available from https://proceedings.mlr.press/v162/majumdar22a.html.

Related Material