# Adaptive debiased machine learning using data-driven model selection techniques

arXiv: https://arxiv.org/abs/2307.12544 · HTML: https://mlsanity.com/paper/2307.12544/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 118 references that are in the map.
- First posted on arXiv 24 Jul 2023.

## Builds on (most cited first, 26 of 26)

- [Quasi-Oracle Estimation of Heterogeneous Treatment Effects](https://mlsanity.com/paper/1712.04912/) (2017)
- [ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R](https://mlsanity.com/paper/1508.04409/) (2015)
- [Controlling the false discovery rate via knockoffs](https://mlsanity.com/paper/1404.5609/) (2014)
- [Exact post-selection inference, with application to the lasso](https://mlsanity.com/paper/1311.6238/) (2013)
- [Confidence Intervals and Hypothesis Testing for High-Dimensional Regression](https://mlsanity.com/paper/1306.3171/) (2013)
- [De-Biased Machine Learning of Global and Local Parameters Using Regularized Riesz Representers](https://mlsanity.com/paper/1802.08667/) (2018)
- [Selective Sequential Model Selection](https://mlsanity.com/paper/1512.02565/) (2015)
- [Efficient Estimation of Pathwise Differentiable Target Parameters with the Undersmoothed Highly Adaptive Lasso](https://mlsanity.com/paper/1908.05607/) (2019)
- [Selective inference for effect modification via the lasso](https://mlsanity.com/paper/1705.08020/) (2017)
- [Multivariate extensions of isotonic regression and total variation denoising via entire monotonicity and Hardy-Krause variation](https://mlsanity.com/paper/1903.01395/) (2019)
- [On Weighted Orthogonal Learners for Heterogeneous Treatment Effects](https://mlsanity.com/paper/2303.12687/) (2023)
- [Selective machine learning of doubly robust functionals](https://mlsanity.com/paper/1911.02029/) (2019)
- [In Defense of the Indefensible: A Very Naive Approach to High-Dimensional Inference](https://mlsanity.com/paper/1705.05543/) (2017)
- [False Discovery Rate Control via Debiased Lasso](https://mlsanity.com/paper/1803.04464/) (2018)
- [Minimax Semiparametric Learning With Approximate Sparsity](https://mlsanity.com/paper/1912.12213/) (2019)
- [On Adaptive Propensity Score Truncation in Causal Inference](https://mlsanity.com/paper/1707.05861/) (2017)
- [MARS via LASSO](https://mlsanity.com/paper/2111.11694/) (2021)
- [Rank-transformed subsampling: inference for multiple data splitting and exchangeable p-values](https://mlsanity.com/paper/2301.02739/) (2023)
- [Universal sieve-based strategies for efficient estimation using machine learning tools](https://mlsanity.com/paper/2003.01856/) (2020)
- [A nonparametric super-efficient estimator of the average treatment effect](https://mlsanity.com/paper/1901.05056/) (2019)
- [The costs and benefits of uniformly valid causal inference with high-dimensional nuisance parameters](https://mlsanity.com/paper/2105.02071/) (2021)
- [Why Machine Learning Cannot Ignore Maximum Likelihood Estimation](https://mlsanity.com/paper/2110.12112/) (2021)
- [Collaborative-controlled LASSO for Constructing Propensity Score-based Estimators in High-Dimensional Data](https://mlsanity.com/paper/1706.10029/) (2017)
- [Nonparametric Bootstrap Inference for the Targeted Highly Adaptive LASSO Estimator](https://mlsanity.com/paper/1905.10299/) (2019)
- [Faithful Variable Screening for High-Dimensional Convex Regression](https://mlsanity.com/paper/1411.1805/) (2014)
- [On Selecting and Conditioning in Multiple Testing and Selective Inference](https://mlsanity.com/paper/2207.13480/) (2022)

## Built on by (most cited first, 7 of 7)

- [Quantifying Aleatoric Uncertainty of the Treatment Effect: A Novel Orthogonal Learner](https://mlsanity.com/paper/2411.03387/) (2024)
- [Adaptive-TMLE for the Average Treatment Effect based on Randomized Controlled Trial Augmented with Real-World Data](https://mlsanity.com/paper/2405.07186/) (2024)
- [Automatic Debiased Machine Learning for Smooth Functionals of Nonparametric M-Estimands](https://mlsanity.com/paper/2501.11868/) (2025)
- [Efficient Inference after Directionally Stable Adaptive Experiments](https://mlsanity.com/paper/2602.21478/) (2026)
- [Efficient Targeted Maximum Likelihood Estimators for Two-Phase Design Problems](https://mlsanity.com/paper/2602.24131/) (2026)
- [Targeted Deep Architectures: A TMLE-Based Framework for Robust Causal Inference in Neural Networks](https://mlsanity.com/paper/2507.12435/) (2025)
- [Improving the Efficiency of Subgroup Analysis in Randomized Controlled Trials with TMLE](https://mlsanity.com/paper/2605.15483/) (2026)

## Abstract (arXiv)

Debiased machine learning estimators for smooth functionals in nonparametric models can exhibit substantial variability and instability, often leading practitioners to instead rely on parametric or semiparametric working models. Such models, however, may be misspecified and can therefore introduce bias. We study how data-driven model selection can be combined with debiased machine learning to construct estimators that adapt to structure in the data-generating distribution. To this end, we propose Adaptive Debiased Machine Learning (ADML), a nonparametric framework for constructing superefficient estimators of pathwise differentiable parameters. The framework unifies a broad class of previously proposed adaptive estimators, including methods based on variable selection, learned feature representations, and collaborative targeted learning. It requires only high-level conditions and approximate validity of the selection procedure, which are implied by lower-level conditions already assumed in important settings, including sieve-based selection, sparsity-based methods such as the Lasso, and data-adaptive feature representations. We show that ADML estimators yield regular and efficient root-(n) inference for an oracle projection parameter induced by a data-adaptive oracle submodel. This oracle parameter coincides with the target parameter at the true distribution but typically has a smaller efficiency bound, thereby yielding superefficiency for the target parameter. As a practical illustration, we introduce a broad class of automatic ADML estimators for continuous linear functionals of the outcome regression, in which model selection is performed directly on the regression itself. Motivated by overlap challenges in causal inference, we develop new superefficient plug-in estimators for the average treatment effect based on calibration in semiparametric regression models.