Debiased Front-Door Learners for Heterogeneous Effects

2025-09-26

AI Summary

This paper addresses the identification of heterogeneous treatment effects (HTE) in observational studies with unobserved confounding of the treatment-outcome relationship, yet where the mediator is unconfounded. We propose two novel debiased learning methods that systematically integrate the front-door adjustment into the debiased machine learning framework—marking the first such application. Our approaches achieve near-oracle convergence rates and maintain robustness even when auxiliary models converge at slower, nonparametric rates. Methodologically, they unify front-door identification, the R-learner, and the DR-learner frameworks, incorporating nonparametric estimation and rigorous error analysis. Extensive evaluation on synthetic data and the real-world Fatality Analysis Reporting System (FARS) dataset—assessing the heterogeneous causal impact of seat belt laws on traffic fatality rates—demonstrates both high accuracy and sample efficiency. The implementation is publicly available.

Technology Category

• Machine Learning: Causal Learning
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

In observational settings where treatment and outcome share unmeasured confounders but an observed mediator remains unconfounded, the front-door (FD) adjustment identifies causal effects through the mediator. We study the heterogeneous treatment effect (HTE) under FD identification and introduce two debiased learners: FD-DR-Learner and FD-R-Learner. Both attain fast, quasi-oracle rates (i.e., performance comparable to an oracle that knows the nuisances) even when nuisance functions converge as slowly as n^-1/4. We provide error analyses establishing debiasedness and demonstrate robust empirical performance in synthetic studies and a real-world case study of primary seat-belt laws using Fatality Analysis Reporting System (FARS) dataset. Together, these results indicate that the proposed learners deliver reliable and sample-efficient HTE estimates in FD scenarios. The implementation is available at https://github.com/yonghanjung/FD-CATE. Keywords: Front-door adjustment; Heterogeneous treatment effects; Debiased learning; Quasi-oracle rates; Causal inference.

Problem

• Estimating heterogeneous treatment effects under front-door identification
• Debiasing causal learners with slow-converging nuisance functions
• Providing reliable HTE estimates in observational studies with unmeasured confounders

Innovation

• Debiased learners for heterogeneous treatment effects
• Front-door adjustment identifies causal effects
• Fast quasi-oracle rates with slow convergence