Causal Inference using Gaussian Processes with Structured Latent Confounders

Sam Witty, Kenta Takatsu, David Jensen, Vikash Mansinghka
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:10313-10323, 2020.

Abstract

Latent confounders—unobserved variables that influence both treatment and outcome—can bias estimates of causal effects. In some cases, these confounders are shared across observations, e.g. all students taking a course are influenced by the course’s difficulty in addition to any educational interventions they receive individually. This paper shows how to semiparametrically model latent confounders that have this structure and thereby improve estimates of causal effects. The key innovations are a hierarchical Bayesian model, Gaussian processes with structured latent confounders (GP-SLC), and a Monte Carlo inference algorithm for this model based on elliptical slice sampling. GP-SLC provides principled Bayesian uncertainty estimates of individual treatment effect with minimal assumptions about the functional forms relating confounders, covariates, treatment, and outcome. Finally, this paper shows GP-SLC is competitive with or more accurate than widely used causal inference techniques on three benchmark datasets, including the Infant Health and Development Program and a dataset showing the effect of changing temperatures on state-wide energy consumption across New England.

Cite this Paper

BibTeX
@InProceedings{pmlr-v119-witty20a, title = {Causal Inference using {G}aussian Processes with Structured Latent Confounders}, author = {Witty, Sam and Takatsu, Kenta and Jensen, David and Mansinghka, Vikash}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {10313--10323}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/witty20a/witty20a.pdf}, url = {https://proceedings.mlr.press/v119/witty20a.html}, abstract = {Latent confounders—unobserved variables that influence both treatment and outcome—can bias estimates of causal effects. In some cases, these confounders are shared across observations, e.g. all students taking a course are influenced by the course’s difficulty in addition to any educational interventions they receive individually. This paper shows how to semiparametrically model latent confounders that have this structure and thereby improve estimates of causal effects. The key innovations are a hierarchical Bayesian model, Gaussian processes with structured latent confounders (GP-SLC), and a Monte Carlo inference algorithm for this model based on elliptical slice sampling. GP-SLC provides principled Bayesian uncertainty estimates of individual treatment effect with minimal assumptions about the functional forms relating confounders, covariates, treatment, and outcome. Finally, this paper shows GP-SLC is competitive with or more accurate than widely used causal inference techniques on three benchmark datasets, including the Infant Health and Development Program and a dataset showing the effect of changing temperatures on state-wide energy consumption across New England.} }
Endnote
%0 Conference Paper %T Causal Inference using Gaussian Processes with Structured Latent Confounders %A Sam Witty %A Kenta Takatsu %A David Jensen %A Vikash Mansinghka %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-witty20a %I PMLR %P 10313--10323 %U https://proceedings.mlr.press/v119/witty20a.html %V 119 %X Latent confounders—unobserved variables that influence both treatment and outcome—can bias estimates of causal effects. In some cases, these confounders are shared across observations, e.g. all students taking a course are influenced by the course’s difficulty in addition to any educational interventions they receive individually. This paper shows how to semiparametrically model latent confounders that have this structure and thereby improve estimates of causal effects. The key innovations are a hierarchical Bayesian model, Gaussian processes with structured latent confounders (GP-SLC), and a Monte Carlo inference algorithm for this model based on elliptical slice sampling. GP-SLC provides principled Bayesian uncertainty estimates of individual treatment effect with minimal assumptions about the functional forms relating confounders, covariates, treatment, and outcome. Finally, this paper shows GP-SLC is competitive with or more accurate than widely used causal inference techniques on three benchmark datasets, including the Infant Health and Development Program and a dataset showing the effect of changing temperatures on state-wide energy consumption across New England.
APA
Witty, S., Takatsu, K., Jensen, D. & Mansinghka, V.. (2020). Causal Inference using Gaussian Processes with Structured Latent Confounders. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:10313-10323 Available from https://proceedings.mlr.press/v119/witty20a.html.

Related Material