Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization

Abstract

We study first-order algorithms that are uniformly stable for empirical risk minimization (ERM) problems that are convex and smooth with respect to $p$-norms, $p \geq 1$. We propose a black-box reduction method that, by employing properties of uniformly convex regularizers, turns an optimization algorithm for H{ö}lder smooth convex losses into a uniformly stable learning algorithm with optimal statistical risk bounds on the excess risk, up to a constant factor depending on $p$. Achieving a black-box reduction for uniform stability was posed as an open question by Attia and Koren (2022), which had solved the Euclidean case $p=2$. We explore applications that leverage non-Euclidean geometry in addressing binary classification problems.

Cite

Text

Vary et al. "Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization." Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 2025.

Markdown

[Vary et al. "Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization." Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 2025.](https://mlanthology.org/aistats/2025/vary2025aistats-blackbox/)

BibTeX

@inproceedings{vary2025aistats-blackbox,
  title     = {{Black-Box Uniform Stability for Non-Euclidean Empirical Risk Minimization}},
  author    = {Vary, Simon and Martínez-Rubio, David and Rebeschini, Patrick},
  booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics},
  year      = {2025},
  pages     = {4942-4950},
  volume    = {258},
  url       = {https://mlanthology.org/aistats/2025/vary2025aistats-blackbox/}
}