Learning Representations for Counterfactual Inference

Fredrik Johansson, Uri Shalit, David Sontag
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:3020-3029, 2016.

Abstract

Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology. We consider the task of answering counterfactual questions such as, “Would this patient have lower blood sugar had she received a different medication?". We propose a new algorithmic framework for counterfactual inference which brings together ideas from domain adaptation and representation learning. In addition to a theoretical justification, we perform an empirical comparison with previous approaches to causal inference from observational data. Our deep learning algorithm significantly outperforms the previous state-of-the-art.

Cite this Paper

BibTeX
@InProceedings{pmlr-v48-johansson16, title = {Learning Representations for Counterfactual Inference}, author = {Johansson, Fredrik and Shalit, Uri and Sontag, David}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {3020--3029}, year = {2016}, editor = {Balcan, Maria Florina and Weinberger, Kilian Q.}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/johansson16.pdf}, url = {https://proceedings.mlr.press/v48/johansson16.html}, abstract = {Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology. We consider the task of answering counterfactual questions such as, “Would this patient have lower blood sugar had she received a different medication?". We propose a new algorithmic framework for counterfactual inference which brings together ideas from domain adaptation and representation learning. In addition to a theoretical justification, we perform an empirical comparison with previous approaches to causal inference from observational data. Our deep learning algorithm significantly outperforms the previous state-of-the-art.} }
Endnote
%0 Conference Paper %T Learning Representations for Counterfactual Inference %A Fredrik Johansson %A Uri Shalit %A David Sontag %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-johansson16 %I PMLR %P 3020--3029 %U https://proceedings.mlr.press/v48/johansson16.html %V 48 %X Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology. We consider the task of answering counterfactual questions such as, “Would this patient have lower blood sugar had she received a different medication?". We propose a new algorithmic framework for counterfactual inference which brings together ideas from domain adaptation and representation learning. In addition to a theoretical justification, we perform an empirical comparison with previous approaches to causal inference from observational data. Our deep learning algorithm significantly outperforms the previous state-of-the-art.
RIS
TY - CPAPER TI - Learning Representations for Counterfactual Inference AU - Fredrik Johansson AU - Uri Shalit AU - David Sontag BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-johansson16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 3020 EP - 3029 L1 - http://proceedings.mlr.press/v48/johansson16.pdf UR - https://proceedings.mlr.press/v48/johansson16.html AB - Observational studies are rising in importance due to the widespread accumulation of data in fields such as healthcare, education, employment and ecology. We consider the task of answering counterfactual questions such as, “Would this patient have lower blood sugar had she received a different medication?". We propose a new algorithmic framework for counterfactual inference which brings together ideas from domain adaptation and representation learning. In addition to a theoretical justification, we perform an empirical comparison with previous approaches to causal inference from observational data. Our deep learning algorithm significantly outperforms the previous state-of-the-art. ER -
APA
Johansson, F., Shalit, U. & Sontag, D.. (2016). Learning Representations for Counterfactual Inference. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:3020-3029 Available from https://proceedings.mlr.press/v48/johansson16.html.

Related Material