Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effects
This work addresses the challenge of evaluating decisions from non-experimental data in fields like healthcare and economics, representing an incremental improvement through theoretical bounds and algorithmic extensions.
The authors tackled the problem of estimating individual-level causal effects from observational data by developing generalization bounds based on treatment group distances and creating representation learning algorithms that minimize these bounds through regularization. Their experimental evaluation on real and synthetic data demonstrated the value of their proposed architecture and regularization scheme.
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.