Double Robust Representation Learning for Counterfactual Prediction
This addresses causal inference challenges in healthcare, policy, and social sciences, offering an incremental improvement over existing methods.
The paper tackles the problem of debiasing causal estimators in high-dimensional observational studies by proposing a scalable method for learning double-robust representations, which ensures consistent causal estimation if either the propensity score or outcome model is correct, and shows competitive performance on real-world and synthetic data.
Causal inference, or counterfactual prediction, is central to decision making in healthcare, policy and social sciences. To de-bias causal estimators with high-dimensional data in observational studies, recent advances suggest the importance of combining machine learning models for both the propensity score and the outcome function. We propose a novel scalable method to learn double-robust representations for counterfactual predictions, leading to consistent causal estimation if the model for either the propensity score or the outcome, but not necessarily both, is correctly specified. Specifically, we use the entropy balancing method to learn the weights that minimize the Jensen-Shannon divergence of the representation between the treated and control groups, based on which we make robust and efficient counterfactual predictions for both individual and average treatment effects. We provide theoretical justifications for the proposed method. The algorithm shows competitive performance with the state-of-the-art on real world and synthetic data.