AutoCATE: End-to-End, Automated Treatment Effect Estimation

Toon Vanderschueren, Tim Verdonck, Mihaela Van Der Schaar, Wouter Verbeke
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:60880-60904, 2025.

Abstract

Estimating causal effects is crucial in domains like healthcare, economics, and education. Despite advances in machine learning (ML) for estimating conditional average treatment effects (CATE), the practical adoption of these methods remains limited, due to the complexities of implementing, tuning, and validating them. To address these challenges, we formalize the search for an optimal ML pipeline for CATE estimation as a counterfactual Combined Algorithm Selection and Hyperparameter (CASH) optimization. We introduce AutoCATE, the first end-to-end, automated solution for CATE estimation. Unlike prior approaches that address only parts of this problem, AutoCATE integrates evaluation, estimation, and ensembling in a unified framework. AutoCATE enables comprehensive comparisons of different protocols, yielding novel insights into CATE estimation and a final configuration that outperforms commonly used strategies. To facilitate broad adoption and further research, we release AutoCATE as an open-source software package.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-vanderschueren25a, title = {{A}uto{CATE}: End-to-End, Automated Treatment Effect Estimation}, author = {Vanderschueren, Toon and Verdonck, Tim and Van Der Schaar, Mihaela and Verbeke, Wouter}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {60880--60904}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/vanderschueren25a/vanderschueren25a.pdf}, url = {https://proceedings.mlr.press/v267/vanderschueren25a.html}, abstract = {Estimating causal effects is crucial in domains like healthcare, economics, and education. Despite advances in machine learning (ML) for estimating conditional average treatment effects (CATE), the practical adoption of these methods remains limited, due to the complexities of implementing, tuning, and validating them. To address these challenges, we formalize the search for an optimal ML pipeline for CATE estimation as a counterfactual Combined Algorithm Selection and Hyperparameter (CASH) optimization. We introduce AutoCATE, the first end-to-end, automated solution for CATE estimation. Unlike prior approaches that address only parts of this problem, AutoCATE integrates evaluation, estimation, and ensembling in a unified framework. AutoCATE enables comprehensive comparisons of different protocols, yielding novel insights into CATE estimation and a final configuration that outperforms commonly used strategies. To facilitate broad adoption and further research, we release AutoCATE as an open-source software package.} }
Endnote
%0 Conference Paper %T AutoCATE: End-to-End, Automated Treatment Effect Estimation %A Toon Vanderschueren %A Tim Verdonck %A Mihaela Van Der Schaar %A Wouter Verbeke %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-vanderschueren25a %I PMLR %P 60880--60904 %U https://proceedings.mlr.press/v267/vanderschueren25a.html %V 267 %X Estimating causal effects is crucial in domains like healthcare, economics, and education. Despite advances in machine learning (ML) for estimating conditional average treatment effects (CATE), the practical adoption of these methods remains limited, due to the complexities of implementing, tuning, and validating them. To address these challenges, we formalize the search for an optimal ML pipeline for CATE estimation as a counterfactual Combined Algorithm Selection and Hyperparameter (CASH) optimization. We introduce AutoCATE, the first end-to-end, automated solution for CATE estimation. Unlike prior approaches that address only parts of this problem, AutoCATE integrates evaluation, estimation, and ensembling in a unified framework. AutoCATE enables comprehensive comparisons of different protocols, yielding novel insights into CATE estimation and a final configuration that outperforms commonly used strategies. To facilitate broad adoption and further research, we release AutoCATE as an open-source software package.
APA
Vanderschueren, T., Verdonck, T., Van Der Schaar, M. & Verbeke, W.. (2025). AutoCATE: End-to-End, Automated Treatment Effect Estimation. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:60880-60904 Available from https://proceedings.mlr.press/v267/vanderschueren25a.html.

Related Material