Treatment Effect Estimation for Exponential Family Outcomes using Neural Networks with Targeted Regularization

TL;DR

This paper extends neural network-based treatment effect estimators to exponential family outcomes using targeted regularization, achieving doubly robust properties and strong theoretical guarantees.

Contribution

It generalizes targeted regularization for neural networks from Gaussian to exponential family outcomes, introducing a new estimator with proven convergence rates.

Findings

The proposed estimator is effective across various exponential family distributions.

Theoretical analysis confirms the estimator's convergence properties.

Experimental results demonstrate improved treatment effect estimation accuracy.

Abstract

Neural Networks (NNs) have became a natural choice for treatment effect estimation due to their strong approximation capabilities. Nevertheless, how to design NN-based estimators with desirable properties, such as low bias and doubly robustness, still remains a significant challenge. A common approach to address this is targeted regularization, which modifies the objective function of NNs. However, existing works on targeted regularization are limited to Gaussian-distributed outcomes, significantly restricting their applicability in real-world scenarios. In this work, we aim to bridge this blank by extending this framework to the boarder exponential family outcomes. Specifically, we first derive the von-Mises expansion of the Average Dose function of Canonical Functions (ADCF), which inspires us how to construct a doubly robust estimator with good properties. Based on this, we develop a NN-based estimator for ADCF by generalizing functional targeted regularization to exponential families, and provide the corresponding theoretical convergence rate. Extensive experimental results demonstrate the effectiveness of our proposed model.