Learning Overlapping Representations for the Estimation of Individualized Treatment Effects

• DOI: 10.48550/arxiv.2001.04754
• Publication Date: 2020-01-01
• Authors: Yao Zhang; Alexis Bellot; Mihaela van der S...
• Keywords: Regularization

The choice of making an intervention depends on its potential benefit or harm in comparison to alternatives. Estimating the likely outcome alternatives from observational data is a challenging problem as all outcomes are never observed, and selection bias precludes direct differently intervened groups. Despite their empirical success, we show that algorithms learn domain-invariant representations inputs (on which make predictions) often inappropriate, develop generalization bounds demonstrate dependence domain overlap highlight need for invertible latent maps. Based these results, deep kernel regression algorithm posterior regularization framework substantially outperforms state-of-the-art variety benchmarks sets.