# Efficient and Robust Semi-supervised Estimation of ATE with Partially Annotated Treatment and Response

arXiv: https://arxiv.org/abs/2110.12336 · HTML: https://mlsanity.com/paper/2110.12336/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Oct 2021.

## Built on by (most cited first, 1 of 1)

- [Continuous Treatment Effects with Surrogate Outcomes](https://mlsanity.com/paper/2402.00168/) (2024)

## Abstract (arXiv)

A notable challenge of leveraging Electronic Health Records (EHR) for treatment effect assessment is the lack of precise information on important clinical variables, including the treatment received and the response. Both treatment information and response often cannot be accurately captured by readily available EHR features and require labor intensive manual chart review to precisely annotate, which limits the number of available gold standard labels on these key variables. We consider average treatment effect (ATE) estimation under such a semi-supervised setting with a large number of unlabeled samples containing both confounders and imperfect EHR features for treatment and response. We derive the efficient influence function for ATE and use it to construct a semi-supervised multiple machine learning (SMMAL) estimator. We showcase that our SMMAL estimator is semi-parametric efficient with B-spline regression under low-dimensional smooth models. We develop the adaptive sparsity/model doubly robust estimation under high-dimensional logistic propensity score and outcome regression models. Results from simulation studies support the validity of our SMMAL method and its superiority over supervised benchmarks.