A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization

Kunjie Ren; Luo Luo

ICML 2025poster0 citations

A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization

Abstract

This paper studies zeroth-order optimization for stochastic convex minimization problems. We propose a parameter-free stochastic zeroth-order method (POEM), which introduces a step-size scheme based on the distance over finite difference and an adaptive smoothing parameter. Our theoretical analysis shows that POEM achieves near-optimal stochastic zeroth-order oracle complexity. Furthermore, numerical experiments demonstrate that POEM outperforms existing zeroth-order methods in practice.

zeroth-order optimizationstochastic convex optimizationparameter-free optimization

BibTeX

@inproceedings{
ren2025a,
title={A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization},
author={Kunjie Ren and Luo Luo},
booktitle={Forty-second International Conference on Machine Learning},
year={2025},
url={https://openreview.net/forum?id=JVouJNs8Vr}
}