mlsanityData to 7 Oct 2026

Orthogonal Machine Learning: Power and Limitations

arXiv 1711.00342 ↗First posted 1 Nov 2017cs.LG econ.EM math.STcited by 28

Not on the map. Why?

Abstract (arXiv)

Double machine learning provides √n-consistent estimates of parameters of interest even when high-dimensional or nonparametric nuisance parameters are estimated at an n-1/4 rate. The key is to employ Neyman-orthogonal moment equations which are first-order insensitive to perturbations in the nuisance parameters. We show that the n-1/4 requirement can be improved to n-1/(2k+2) by employing a k-th order notion of orthogonality that grants robustness to more complex or higher-dimensional nuisance parameters. In the partially linear regression setting popular in causal inference, we show that we can construct second-order orthogonal moments if and only if the treatment residual is not normally distributed. Our proof relies on Stein's lemma and may be of independent interest. We conclude by demonstrating the robustness benefits of an explicit doubly-orthogonal estimation procedure for treatment effect.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). First posted on arXiv 1 Nov 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 7; 2023: 5; 2024: 3; 2025: 3; 2026: 5.

28 papers.

Sort:

  1. 2021Outside the Echo Chamber: Optimizing the Performative Riskcited by 95
  2. 2023Multi-Task Reinforcement Learning with Mixture of Orthogonal Expertscited by 45
  3. 2021Deeply-Debiased Off-Policy Interval Estimationcited by 36
  4. 2024Applied Causal Inference Powered by ML and AIcited by 30
  5. 2021Doubly-Valid/Doubly-Sharp Sensitivity Analysis for Causal Inference with Unmeasured Confoundingcited by 28
  6. 2021DoubleML -- An Object-Oriented Implementation of Double Machine Learning in Pythoncited by 23
  7. 2023Mitigating Adversarial Vulnerability through Causal Parameter Estimation by Adversarial Double Machine Learningcited by 11
  8. 2022Partial Identification with Noisy Covariates: A Robust Optimization Approachcited by 10
  9. 2022Off-policy estimation of linear functionals: Non-asymptotic theory for semi-parametric efficiencycited by 10
  10. 2024Learning Decision Policies with Instrumental Variables through Double Machine Learningcited by 5
  11. 2025Differentially Private Learners for Heterogeneous Treatment Effectscited by 5
  12. 2022Coordinated Double Machine Learningcited by 4
16 more
  1. 2024Language Models as Causal Effect Generatorscited by 4
  2. 2023Near Optimal Heteroscedastic Regression with Symbiotic Learningcited by 3
  3. 2022Orthogonal Statistical Learning with Self-Concordant Losscited by 2
  4. 2022Robust Causal Learning for the Estimation of Average Treatment Effectscited by 2
  5. 2023Covariance alignment: from maximum likelihood estimation to Gromov-Wassersteincited by 1
  6. 2026De-paradox Tree: Breaking Down Simpson's Paradox via A Kernel-Based Partition Algorithmcited by 1
  7. 2021Robust Orthogonal Machine Learning of Treatment Effectscited by 0
  8. 2022Machine Learning based Framework for Robust Price-Sensitivity Estimation with Application to Airline Pricingcited by 0
  9. 2022Semantic Autoencoder and Its Potential Usage for Adversarial Attackcited by 0
  10. 2023Stable Probability Weighting: Large-Sample and Finite-Sample Estimation and Inference Methods for Heterogeneous Causal Effects of Multivalued Treatments Under Limited Overlapcited by 0
  11. 2025Estimating Treatment Effects with Independent Component Analysiscited by 0
  12. 2025The Unified Non-Convex Framework for Robust Causal Inference: Overcoming the Gaussian Barrier and Optimization Fragilitycited by 0
  13. 2026MEC: Machine-Learning-Assisted Generalized Entropy Calibration for Semi-Supervised Mean Estimationcited by 0
  14. 2026Adaptive Estimation and Inference in Semi-parametric Heterogeneous Clustered Multitask Learning via Neyman Orthogonalitycited by 0
  15. 2026ORTHOBO: Orthogonal Bayesian Hyperparameter Optimizationcited by 0
  16. 2026Sharp Structure-Agnostic Minimax Risk for Partial Linear Modelscited by 0