On the Effects of Irrelevant Variables in Treatment Effect Estimation with Deep Disentanglement

TL;DR

This paper introduces a deep disentanglement method that explicitly identifies and isolates irrelevant variables in treatment effect estimation, improving prediction accuracy and robustness against irrelevant data.

Contribution

We propose a novel deep embedding approach with an autoencoder and orthogonalization to explicitly separate irrelevant variables from relevant factors in treatment effect models.

Findings

Better identification of irrelevant variables in treatment effect models

More accurate treatment effect predictions on synthetic and real datasets

Reduced prediction degradation with added irrelevant variables

Abstract

Estimating treatment effects from observational data is paramount in healthcare, education, and economics, but current deep disentanglement-based methods to address selection bias are insufficiently handling irrelevant variables. We demonstrate in experiments that this leads to prediction errors. We disentangle pre-treatment variables with a deep embedding method and explicitly identify and represent irrelevant variables, additionally to instrumental, confounding and adjustment latent factors. To this end, we introduce a reconstruction objective and create an embedding space for irrelevant variables using an attached autoencoder. Instead of relying on serendipitous suppression of irrelevant variables as in previous deep disentanglement approaches, we explicitly force irrelevant variables into this embedding space and employ orthogonalization to prevent irrelevant information from leaking into the latent space representations of the other factors. Our experiments with synthetic and real-world benchmark datasets show that we can better identify irrelevant variables and more precisely predict treatment effects than previous methods, while prediction quality degrades less when additional irrelevant variables are introduced.