DRCFS: Doubly Robust Causal Feature Selection

Francesco Quinzan, Ashkan Soleymani, Patrick Jaillet, Cristian R. Rojas, Stefan Bauer
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:28468-28491, 2023.

Abstract

Knowing the features of a complex system that are highly relevant to a particular target variable is of fundamental interest in many areas of science. Existing approaches are often limited to linear settings, sometimes lack guarantees, and in most cases, do not scale to the problem at hand, in particular to images. We propose DRCFS, a doubly robust feature selection method for identifying the causal features even in nonlinear and high dimensional settings. We provide theoretical guarantees, illustrate necessary conditions for our assumptions, and perform extensive experiments across a wide range of simulated and semi-synthetic datasets. DRCFS significantly outperforms existing state-of-the-art methods, selecting robust features even in challenging highly non-linear and high-dimensional problems.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-quinzan23a, title = {{DRCFS}: Doubly Robust Causal Feature Selection}, author = {Quinzan, Francesco and Soleymani, Ashkan and Jaillet, Patrick and Rojas, Cristian R. and Bauer, Stefan}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {28468--28491}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/quinzan23a/quinzan23a.pdf}, url = {https://proceedings.mlr.press/v202/quinzan23a.html}, abstract = {Knowing the features of a complex system that are highly relevant to a particular target variable is of fundamental interest in many areas of science. Existing approaches are often limited to linear settings, sometimes lack guarantees, and in most cases, do not scale to the problem at hand, in particular to images. We propose DRCFS, a doubly robust feature selection method for identifying the causal features even in nonlinear and high dimensional settings. We provide theoretical guarantees, illustrate necessary conditions for our assumptions, and perform extensive experiments across a wide range of simulated and semi-synthetic datasets. DRCFS significantly outperforms existing state-of-the-art methods, selecting robust features even in challenging highly non-linear and high-dimensional problems.} }
Endnote
%0 Conference Paper %T DRCFS: Doubly Robust Causal Feature Selection %A Francesco Quinzan %A Ashkan Soleymani %A Patrick Jaillet %A Cristian R. Rojas %A Stefan Bauer %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-quinzan23a %I PMLR %P 28468--28491 %U https://proceedings.mlr.press/v202/quinzan23a.html %V 202 %X Knowing the features of a complex system that are highly relevant to a particular target variable is of fundamental interest in many areas of science. Existing approaches are often limited to linear settings, sometimes lack guarantees, and in most cases, do not scale to the problem at hand, in particular to images. We propose DRCFS, a doubly robust feature selection method for identifying the causal features even in nonlinear and high dimensional settings. We provide theoretical guarantees, illustrate necessary conditions for our assumptions, and perform extensive experiments across a wide range of simulated and semi-synthetic datasets. DRCFS significantly outperforms existing state-of-the-art methods, selecting robust features even in challenging highly non-linear and high-dimensional problems.
APA
Quinzan, F., Soleymani, A., Jaillet, P., Rojas, C.R. & Bauer, S.. (2023). DRCFS: Doubly Robust Causal Feature Selection. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:28468-28491 Available from https://proceedings.mlr.press/v202/quinzan23a.html.

Related Material