Overlap-Adaptive Regularization for Conditional Average Treatment Effect Estimation
Not on the map. Why?
Abstract (arXiv)
The conditional average treatment effect (CATE) is widely used in personalized medicine to inform therapeutic decisions. However, state-of-the-art methods for CATE estimation (so-called meta-learners) often perform poorly in the presence of low overlap. In this work, we introduce a new approach to tackle this issue and improve the performance of existing meta-learners in the low-overlap regions. Specifically, we introduce Overlap-Adaptive Regularization (OAR) that regularizes target models proportionally to overlap weights so that, informally, the regularization is higher in regions with low overlap. To the best of our knowledge, our OAR is the first approach to leverage overlap weights in the regularization terms of the meta-learners. Our OAR approach is flexible and works with any existing CATE meta-learner: we demonstrate how OAR can be applied to both parametric and non-parametric second-stage models. Furthermore, we propose debiased versions of our OAR that preserve the Neyman-orthogonality of existing meta-learners and thus ensure more robust inference. Through a series of (semi-)synthetic experiments, we demonstrate that our OAR significantly improves CATE estimation in low-overlap settings in comparison to constant regularization.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 39 of its 69 references that are in the map. First posted on arXiv 29 Sep 2025.
Builds on
39 of 69 references.
References · 39
- 2015Estimation and Inference of Heterogeneous Treatment Effects using Random Forestscited by 600
- 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 478
- 2017Quasi-Oracle Estimation of Heterogeneous Treatment Effectscited by 332
- 2016Learning Representations for Counterfactual Inferencecited by 329
- 2016Generalized Random Forestscited by 312
- 2017Nearly-tight VC-dimension and pseudodimension bounds for piecewise linear neural networkscited by 311
- 2020How Does Mixup Help With Robustness and Generalization?cited by 193
- 2017Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processescited by 158
- 2019Orthogonal Statistical Learningcited by 151
- 2021Nonparametric Estimation of Heterogeneous Treatment Effects: From Theory to Learning Algorithmscited by 144
- 2022Semiparametric doubly robust targeted double machine learning: a reviewcited by 133
- 2017Automated versus do-it-yourself methods for causal inference: Lessons learned from a data analysis competitioncited by 106
27 more
- 2013Dropout Training as Adaptive Regularizationcited by 93
- 2017Double/Debiased/Neyman Machine Learning of Treatment Effectscited by 76
- 2021On Inductive Biases for Heterogeneous Treatment Effect Estimationcited by 75
- 2024Causal machine learning for predicting treatment outcomescited by 74
- 2020Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effectscited by 72
- 2020The Implicit and Explicit Regularization Effects of Dropoutcited by 71
- 2016Balancing Covariates via Propensity Score Weightingcited by 65
- 2019Estimating Treatment Effects with Causal Forests: An Applicationcited by 58
- 2021Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confoundingcited by 50
- 2020Counterfactual Representation Learning with Balancing Weightscited by 45
- 2020Explicit Regularisation in Gaussian Noise Injectionscited by 37
- 2023In Search of Insights, Not Magic Bullets: Towards Demystification of the Model Selection Dilemma in Heterogeneous Treatment Effect Estimationcited by 29
- 2020Dropout: Explicit Forms and Capacity Controlcited by 27
- 2019The Implicit Regularization of Ordinary Least Squares Ensemblescited by 25
- 2019Noise Regularization for Conditional Density Estimationcited by 23
- 2018On the Implicit Bias of Dropoutcited by 22
- 2022Normalizing Flows for Interventional Density Estimationcited by 21
- 2023On the Optimality of Misspecified Kernel Ridge Regressioncited by 20
- 2023Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimationcited by 20
- 2024Model-agnostic meta-learners for estimating heterogeneous treatment effects over timecited by 19
- 2017Bayesian Nonparametric Causal Inference: Information Rates and Learning Algorithmscited by 16
- 2019On Dropout and Nuclear Norm Regularizationcited by 12
- 2024Simplifying debiased inference via automatic differentiation and probabilistic programmingcited by 10
- 2018Noisin: Unbiased Regularization for Recurrent Neural Networkscited by 8
- 2021Structured Dropout Variational Inference for Bayesian Neural Networkscited by 7
- 2025Orthogonal Representation Learning for Estimating Causal Quantitiescited by 5
- 2022Overlap, matching, or entropy weights: what are we weighting for?cited by 3
Built on by
Citing papers per year: 2025: 1.
1 paper.