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

Kunjie Ren, Luo Luo
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:51450-51470, 2025.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-ren25c, title = {A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization}, author = {Ren, Kunjie and Luo, Luo}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {51450--51470}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/ren25c/ren25c.pdf}, url = {https://proceedings.mlr.press/v267/ren25c.html}, 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.} }
Endnote
%0 Conference Paper %T A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization %A Kunjie Ren %A Luo Luo %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-ren25c %I PMLR %P 51450--51470 %U https://proceedings.mlr.press/v267/ren25c.html %V 267 %X 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.
APA
Ren, K. & Luo, L.. (2025). A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:51450-51470 Available from https://proceedings.mlr.press/v267/ren25c.html.

Related Material