# Adversarial Balancing for Causal Inference

arXiv: https://arxiv.org/abs/1810.07406 · HTML: https://mlsanity.com/paper/1810.07406/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Oct 2018.

## Built on by (most cited first, 10 of 10)

- [A Causal Lens for Controllable Text Generation](https://mlsanity.com/paper/2201.09119/) (2022)
- [Deep Causal Learning: Representation, Discovery and Inference](https://mlsanity.com/paper/2211.03374/) (2022)
- [Continuous Treatment Effect Estimation Using Gradient Interpolation and Kernel Smoothing](https://mlsanity.com/paper/2401.15447/) (2024)
- [Preference Learning for AI Alignment: a Causal Perspective](https://mlsanity.com/paper/2506.05967/) (2025)
- [Outcome Assumptions and Duality Theory for Balancing Weights](https://mlsanity.com/paper/2203.09557/) (2022)
- [PairNet: Training with Observed Pairs to Estimate Individual Treatment Effect](https://mlsanity.com/paper/2406.03864/) (2024)
- [Towards Representation Learning for Weighting Problems in Design-Based Causal Inference](https://mlsanity.com/paper/2409.16407/) (2024)
- [Adversarial Distribution Balancing for Counterfactual Reasoning](https://mlsanity.com/paper/2311.16616/) (2023)
- [Leveraging a Simulator for Learning Causal Representations from Post-Treatment Covariates for CATE](https://mlsanity.com/paper/2502.05037/) (2025)
- [Supervised Learning Objectives for High-Dimensional Treatment Effects](https://mlsanity.com/paper/2605.27281/) (2026)

## Abstract (arXiv)

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.