# Orthogonal Statistical Learning with Self-Concordant Loss

arXiv: https://arxiv.org/abs/2205.00350 · HTML: https://mlsanity.com/paper/2205.00350/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 34 references that are in the map.
- First posted on arXiv 30 Apr 2022.

## Builds on (most cited first, 5 of 5)

- [Orthogonal Statistical Learning](https://mlsanity.com/paper/1901.09036/) (2019)
- [Finite-sample analysis of M-estimators using self-concordance](https://mlsanity.com/paper/1810.06838/) (2018)
- [Orthogonal Machine Learning: Power and Limitations](https://mlsanity.com/paper/1711.00342/) (2017)
- [Regularized Orthogonal Machine Learning for Nonlinear Semiparametric Models](https://mlsanity.com/paper/1806.04823/) (2018)
- [Double/Debiased Machine Learning for Logistic Partially Linear Model](https://mlsanity.com/paper/2009.14461/) (2020)

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

- [Near Optimal Heteroscedastic Regression with Symbiotic Learning](https://mlsanity.com/paper/2306.14288/) (2023)
- [Confidence Sets under Generalized Self-Concordance](https://mlsanity.com/paper/2301.00260/) (2022)

## Abstract (arXiv)

Orthogonal statistical learning and double machine learning have emerged as general frameworks for two-stage statistical prediction in the presence of a nuisance component. We establish non-asymptotic bounds on the excess risk of orthogonal statistical learning methods with a loss function satisfying a self-concordance property. Our bounds improve upon existing bounds by a dimension factor while lifting the assumption of strong convexity. We illustrate the results with examples from multiple treatment effect estimation and generalized partially linear modeling.