# Coordinated Double Machine Learning

arXiv: https://arxiv.org/abs/2206.00885 · HTML: https://mlsanity.com/paper/2206.00885/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 27 references that are in the map.
- First posted on arXiv 2 Jun 2022.

## Builds on (most cited first, 9 of 9)

- [Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes](https://mlsanity.com/paper/1908.08526/) (2019)
- [Orthogonal Machine Learning: Power and Limitations](https://mlsanity.com/paper/1711.00342/) (2017)
- [Double/Debiased Machine Learning for Dynamic Treatment Effects via g-Estimation](https://mlsanity.com/paper/2002.07285/) (2020)
- [DoubleML -- An Object-Oriented Implementation of Double Machine Learning in Python](https://mlsanity.com/paper/2104.03220/) (2021)
- [Distributed Double Machine Learning with a Serverless Architecture](https://mlsanity.com/paper/2101.04025/) (2021)
- [Transformation Models in High-Dimensions](https://mlsanity.com/paper/1712.07364/) (2017)
- [Uniform Inference in High-Dimensional Gaussian Graphical Models](https://mlsanity.com/paper/1808.10532/) (2018)
- [Estimation and Uniform Inference in Sparse High-Dimensional Additive Models](https://mlsanity.com/paper/2004.01623/) (2020)
- [The Bias-Variance Tradeoff of Doubly Robust Estimator with Targeted L1 regularized Neural Networks Predictions](https://mlsanity.com/paper/2108.00990/) (2021)

## Built on by (most cited first, 4 of 4)

- [Mitigating Adversarial Vulnerability through Causal Parameter Estimation by Adversarial Double Machine Learning](https://mlsanity.com/paper/2307.07250/) (2023)
- [Near Optimal Heteroscedastic Regression with Symbiotic Learning](https://mlsanity.com/paper/2306.14288/) (2023)
- [Adaptive Estimation and Inference in Semi-parametric Heterogeneous Clustered Multitask Learning via Neyman Orthogonality](https://mlsanity.com/paper/2605.01907/) (2026)
- [Disentangled Double Machine Learning for Accurate Causal Effect Estimation](https://mlsanity.com/paper/2605.24808/) (2026)

## Abstract (arXiv)

Double machine learning is a statistical method for leveraging complex black-box models to construct approximately unbiased treatment effect estimates given observational data with high-dimensional covariates, under the assumption of a partially linear model. The idea is to first fit on a subset of the samples two non-linear predictive models, one for the continuous outcome of interest and one for the observed treatment, and then to estimate a linear coefficient for the treatment using the remaining samples through a simple orthogonalized regression. While this methodology is flexible and can accommodate arbitrary predictive models, typically trained independently of one another, this paper argues that a carefully coordinated learning algorithm for deep neural networks may reduce the estimation bias. The improved empirical performance of the proposed method is demonstrated through numerical experiments on both simulated and real data.