Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effects

TL;DR

This paper develops theoretical bounds for estimating individual causal effects from observational data, introduces representation learning algorithms to minimize these bounds, and demonstrates their effectiveness through experiments.

Contribution

It provides the first generalization bounds for causal effect estimation based on group distance measures and proposes novel representation learning methods to improve estimation accuracy.

Findings

Theoretical bounds relate to group distance measures and domain adaptation.

Representation learning algorithms effectively reduce estimation error.

Experimental results show improved causal effect estimation with proposed methods.

Abstract

Practitioners in diverse fields such as healthcare, economics and education are eager to apply machine learning to improve decision making. The cost and impracticality of performing experiments and a recent monumental increase in electronic record keeping has brought attention to the problem of evaluating decisions based on non-experimental observational data. This is the setting of this work. In particular, we study estimation of individual-level causal effects, such as a single patient's response to alternative medication, from recorded contexts, decisions and outcomes. We give generalization bounds on the error in estimated effects based on distance measures between groups receiving different treatments, allowing for sample re-weighting. We provide conditions under which our bound is tight and show how it relates to results for unsupervised domain adaptation. Led by our theoretical results, we devise representation learning algorithms that minimize our bound, by regularizing the representation's induced treatment group distance, and encourage sharing of information between treatment groups. We extend these algorithms to simultaneously learn a weighted representation to further reduce treatment group distances. Finally, an experimental evaluation on real and synthetic data shows the value of our proposed representation architecture and regularization scheme.