Adversarial Balancing for Causal Inference

TL;DR

This paper introduces an adversarial reweighting method for causal inference that uses bi-level optimization and GAN-inspired principles to better balance treatment groups and reduce bias in observational data.

Contribution

It proposes a novel adversarial reweighting approach employing bi-level optimization and GAN concepts to improve causal effect estimation from observational data.

Findings

Outperforms previous reweighting methods on benchmark datasets.

Theoretical bounds relate estimation error to discrepancy and weight variability.

Effective in reducing bias and improving causal inference accuracy.

Abstract

Biases in observational data of treatments pose a major challenge to estimating expected treatment outcomes in different populations. An important technique that accounts for these biases is reweighting samples to minimize the discrepancy between treatment groups. We present a novel reweighting approach that uses bi-level optimization to alternately train a discriminator to minimize classification error, and a balancing weights generator that uses exponentiated gradient descent to maximize this error. This approach borrows principles from generative adversarial networks (GANs) to exploit the power of classifiers for measuring two-sample divergence. We provide theoretical results for conditions in which the estimation error is bounded by two factors: (i) the discrepancy measure induced by the discriminator; and (ii) the weights variability. Experimental results on several benchmarks comparing to previous state-of-the-art reweighting methods demonstrate the effectiveness of this approach in estimating causal effects.