# Deconfounding Scores and Representation Learning for Causal Effect Estimation with Weak Overlap

arXiv: https://arxiv.org/abs/2604.00811 · HTML: https://mlsanity.com/paper/2604.00811/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 66 references that are in the map.
- First posted on arXiv 1 Apr 2026.

## Builds on (most cited first, 11 of 11)

- [Increasing the efficiency of randomized trial estimates via linear adjustment for a prognostic score](https://mlsanity.com/paper/2012.09935/) (2020)
- [Risk ratio, odds ratio, risk difference... Which causal measure is easier to generalize?](https://mlsanity.com/paper/2303.16008/) (2023)
- [Off-Policy Evaluation in Markov Decision Processes under Weak Distributional Overlap](https://mlsanity.com/paper/2402.08201/) (2024)
- [Orthogonal Representation Learning for Estimating Causal Quantities](https://mlsanity.com/paper/2502.04274/) (2025)
- [Off-policy evaluation beyond overlap: partial identification through smoothness](https://mlsanity.com/paper/2305.11812/) (2023)
- [Bounding Treatment Effects by Pooling Limited Information across Observations](https://mlsanity.com/paper/2111.05243/) (2021)
- [Improving efficiency in transporting average treatment effects](https://mlsanity.com/paper/2304.00117/) (2023)
- [Towards Representation Learning for Weighting Problems in Design-Based Causal Inference](https://mlsanity.com/paper/2409.16407/) (2024)
- [Bayesian analysis of matrix data with rstiefel](https://mlsanity.com/paper/1304.3673/) (2013)
- [Review of Probability Distributions for Modeling Count Data](https://mlsanity.com/paper/2001.04343/) (2020)
- [A framework for causal inference in the presence of extreme inverse probability weights: the role of overlap weights](https://mlsanity.com/paper/2011.01388/) (2020)

## Built on by (most cited first, 2 of 2)

- [When AI Generates Covariates: Causal Typing and Estimand Drift in Sequential Experiments](https://mlsanity.com/paper/2609.17772/) (2026)
- [Information Set Emulation: Causal Certificates for AI Derived EHR Features](https://mlsanity.com/paper/2609.17777/) (2026)

## Abstract (arXiv)

Overlap, also known as positivity, is a key condition for causal treatment effect estimation. Many popular estimators suffer from high variance and become brittle when features differ strongly across treatment groups. This is especially challenging in high dimensions: the curse of dimensionality can make overlap implausible. To address this, we propose a class of feature representations called deconfounding scores, which preserve both identification and the target of estimation; the classical propensity and prognostic scores are two special cases. We characterize the problem of finding a representation with better overlap as minimizing an overlap divergence under a deconfounding score constraint. We then derive closed-form expressions for a class of deconfounding scores under a broad family of generalized linear models with Gaussian features and show that prognostic scores are overlap-optimal within this class. We conduct extensive experiments to assess this behavior empirically.