Comparison of meta-learners for estimating multi-valued treatment heterogeneous effects

Naoufal Acharki, Ramiro Lugo, Antoine Bertoncello, Josselin Garnier
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:91-132, 2023.

Abstract

Conditional Average Treatment Effects (CATE) estimation is one of the main challenges in causal inference with observational data. In addition to Machine Learning based-models, nonparametric estimators called meta-learners have been developed to estimate the CATE with the main advantage of not restraining the estimation to a specific supervised learning method. This task becomes, however, more complicated when the treatment is not binary as some limitations of the naive extensions emerge. This paper looks into meta-learners for estimating the heterogeneous effects of multi-valued treatments. We consider different meta-learners, and we carry out a theoretical analysis of their error upper bounds as functions of important parameters such as the number of treatment levels, showing that the naive extensions do not always provide satisfactory results. We introduce and discuss meta-learners that perform well as the number of treatments increases. We empirically confirm the strengths and weaknesses of those methods with synthetic and semi-synthetic datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-acharki23a, title = {Comparison of meta-learners for estimating multi-valued treatment heterogeneous effects}, author = {Acharki, Naoufal and Lugo, Ramiro and Bertoncello, Antoine and Garnier, Josselin}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {91--132}, 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/acharki23a/acharki23a.pdf}, url = {https://proceedings.mlr.press/v202/acharki23a.html}, abstract = {Conditional Average Treatment Effects (CATE) estimation is one of the main challenges in causal inference with observational data. In addition to Machine Learning based-models, nonparametric estimators called meta-learners have been developed to estimate the CATE with the main advantage of not restraining the estimation to a specific supervised learning method. This task becomes, however, more complicated when the treatment is not binary as some limitations of the naive extensions emerge. This paper looks into meta-learners for estimating the heterogeneous effects of multi-valued treatments. We consider different meta-learners, and we carry out a theoretical analysis of their error upper bounds as functions of important parameters such as the number of treatment levels, showing that the naive extensions do not always provide satisfactory results. We introduce and discuss meta-learners that perform well as the number of treatments increases. We empirically confirm the strengths and weaknesses of those methods with synthetic and semi-synthetic datasets.} }
Endnote
%0 Conference Paper %T Comparison of meta-learners for estimating multi-valued treatment heterogeneous effects %A Naoufal Acharki %A Ramiro Lugo %A Antoine Bertoncello %A Josselin Garnier %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-acharki23a %I PMLR %P 91--132 %U https://proceedings.mlr.press/v202/acharki23a.html %V 202 %X Conditional Average Treatment Effects (CATE) estimation is one of the main challenges in causal inference with observational data. In addition to Machine Learning based-models, nonparametric estimators called meta-learners have been developed to estimate the CATE with the main advantage of not restraining the estimation to a specific supervised learning method. This task becomes, however, more complicated when the treatment is not binary as some limitations of the naive extensions emerge. This paper looks into meta-learners for estimating the heterogeneous effects of multi-valued treatments. We consider different meta-learners, and we carry out a theoretical analysis of their error upper bounds as functions of important parameters such as the number of treatment levels, showing that the naive extensions do not always provide satisfactory results. We introduce and discuss meta-learners that perform well as the number of treatments increases. We empirically confirm the strengths and weaknesses of those methods with synthetic and semi-synthetic datasets.
APA
Acharki, N., Lugo, R., Bertoncello, A. & Garnier, J.. (2023). Comparison of meta-learners for estimating multi-valued treatment heterogeneous effects. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:91-132 Available from https://proceedings.mlr.press/v202/acharki23a.html.

Related Material