Neural score matching for high-dimensional causal inference

Oscar Clivio, Fabian Falck, Brieuc Lehmann, George Deligiannidis, Chris Holmes
Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, PMLR 151:7076-7110, 2022.

Abstract

Traditional methods for matching in causal inference are impractical for high-dimensional datasets. They suffer from the curse of dimensionality: exact matching and coarsened exact matching find exponentially fewer matches as the input dimension grows, and propensity score matching may match highly unrelated units together. To overcome this problem, we develop theoretical results which motivate the use of neural networks to obtain non-trivial, multivariate balancing scores of a chosen level of coarseness, in contrast to the classical, scalar propensity score. We leverage these balancing scores to perform matching for high-dimensional causal inference and call this procedure neural score matching. We show that our method is competitive against other matching approaches on semi-synthetic high-dimensional datasets, both in terms of treatment effect estimation and reducing imbalance.

Cite this Paper

BibTeX
@InProceedings{pmlr-v151-clivio22a, title = { Neural score matching for high-dimensional causal inference }, author = {Clivio, Oscar and Falck, Fabian and Lehmann, Brieuc and Deligiannidis, George and Holmes, Chris}, booktitle = {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics}, pages = {7076--7110}, year = {2022}, editor = {Camps-Valls, Gustau and Ruiz, Francisco J. R. and Valera, Isabel}, volume = {151}, series = {Proceedings of Machine Learning Research}, month = {28--30 Mar}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v151/clivio22a/clivio22a.pdf}, url = {https://proceedings.mlr.press/v151/clivio22a.html}, abstract = { Traditional methods for matching in causal inference are impractical for high-dimensional datasets. They suffer from the curse of dimensionality: exact matching and coarsened exact matching find exponentially fewer matches as the input dimension grows, and propensity score matching may match highly unrelated units together. To overcome this problem, we develop theoretical results which motivate the use of neural networks to obtain non-trivial, multivariate balancing scores of a chosen level of coarseness, in contrast to the classical, scalar propensity score. We leverage these balancing scores to perform matching for high-dimensional causal inference and call this procedure neural score matching. We show that our method is competitive against other matching approaches on semi-synthetic high-dimensional datasets, both in terms of treatment effect estimation and reducing imbalance. } }
Endnote
%0 Conference Paper %T Neural score matching for high-dimensional causal inference %A Oscar Clivio %A Fabian Falck %A Brieuc Lehmann %A George Deligiannidis %A Chris Holmes %B Proceedings of The 25th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2022 %E Gustau Camps-Valls %E Francisco J. R. Ruiz %E Isabel Valera %F pmlr-v151-clivio22a %I PMLR %P 7076--7110 %U https://proceedings.mlr.press/v151/clivio22a.html %V 151 %X Traditional methods for matching in causal inference are impractical for high-dimensional datasets. They suffer from the curse of dimensionality: exact matching and coarsened exact matching find exponentially fewer matches as the input dimension grows, and propensity score matching may match highly unrelated units together. To overcome this problem, we develop theoretical results which motivate the use of neural networks to obtain non-trivial, multivariate balancing scores of a chosen level of coarseness, in contrast to the classical, scalar propensity score. We leverage these balancing scores to perform matching for high-dimensional causal inference and call this procedure neural score matching. We show that our method is competitive against other matching approaches on semi-synthetic high-dimensional datasets, both in terms of treatment effect estimation and reducing imbalance.
APA
Clivio, O., Falck, F., Lehmann, B., Deligiannidis, G. & Holmes, C.. (2022). Neural score matching for high-dimensional causal inference . Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 151:7076-7110 Available from https://proceedings.mlr.press/v151/clivio22a.html.

Related Material