Coordinated Double Machine Learning

Nitai Fingerhut, Matteo Sesia, Yaniv Romano
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:6499-6513, 2022.

Abstract

Double machine learning is a statistical method for leveraging complex black-box models to construct approximately unbiased treatment effect estimates given observational data with high-dimensional covariates, under the assumption of a partially linear model. The idea is to first fit on a subset of the samples two non-linear predictive models, one for the continuous outcome of interest and one for the observed treatment, and then to estimate a linear coefficient for the treatment using the remaining samples through a simple orthogonalized regression. While this methodology is flexible and can accommodate arbitrary predictive models, typically trained independently of one another, this paper argues that a carefully coordinated learning algorithm for deep neural networks may reduce the estimation bias. The improved empirical performance of the proposed method is demonstrated through numerical experiments on both simulated and real data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-fingerhut22a, title = {Coordinated Double Machine Learning}, author = {Fingerhut, Nitai and Sesia, Matteo and Romano, Yaniv}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {6499--6513}, 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/fingerhut22a/fingerhut22a.pdf}, url = {https://proceedings.mlr.press/v162/fingerhut22a.html}, abstract = {Double machine learning is a statistical method for leveraging complex black-box models to construct approximately unbiased treatment effect estimates given observational data with high-dimensional covariates, under the assumption of a partially linear model. The idea is to first fit on a subset of the samples two non-linear predictive models, one for the continuous outcome of interest and one for the observed treatment, and then to estimate a linear coefficient for the treatment using the remaining samples through a simple orthogonalized regression. While this methodology is flexible and can accommodate arbitrary predictive models, typically trained independently of one another, this paper argues that a carefully coordinated learning algorithm for deep neural networks may reduce the estimation bias. The improved empirical performance of the proposed method is demonstrated through numerical experiments on both simulated and real data.} }
Endnote
%0 Conference Paper %T Coordinated Double Machine Learning %A Nitai Fingerhut %A Matteo Sesia %A Yaniv Romano %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-fingerhut22a %I PMLR %P 6499--6513 %U https://proceedings.mlr.press/v162/fingerhut22a.html %V 162 %X Double machine learning is a statistical method for leveraging complex black-box models to construct approximately unbiased treatment effect estimates given observational data with high-dimensional covariates, under the assumption of a partially linear model. The idea is to first fit on a subset of the samples two non-linear predictive models, one for the continuous outcome of interest and one for the observed treatment, and then to estimate a linear coefficient for the treatment using the remaining samples through a simple orthogonalized regression. While this methodology is flexible and can accommodate arbitrary predictive models, typically trained independently of one another, this paper argues that a carefully coordinated learning algorithm for deep neural networks may reduce the estimation bias. The improved empirical performance of the proposed method is demonstrated through numerical experiments on both simulated and real data.
APA
Fingerhut, N., Sesia, M. & Romano, Y.. (2022). Coordinated Double Machine Learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:6499-6513 Available from https://proceedings.mlr.press/v162/fingerhut22a.html.

Related Material