Fully Adaptive Zeroth-Order Method for Minimizing Functions with Compressible Gradients

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Grapiglia, Geovani Nunes, McKenzie, Daniel
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915389568450560
author Grapiglia, Geovani Nunes
McKenzie, Daniel
author_facet Grapiglia, Geovani Nunes
McKenzie, Daniel
contents We propose an adaptive zeroth-order method for minimizing differentiable functions with $L$-Lipschitz continuous gradients. The method is designed to take advantage of the eventual compressibility of the gradient of the objective function, but it does not require knowledge of the approximate sparsity level $s$ or the Lipschitz constant $L$ of the gradient. We show that the new method performs no more than $O\left(n^{2}ε^{-2}\right)$ function evaluations to find an $ε$-approximate stationary point of an objective function with $n$ variables. Assuming additionally that the gradients of the objective function are compressible, we obtain an improved complexity bound of $O\left(s\log\left(n\right)ε^{-2}\right)$ function evaluations, which holds with high probability. Preliminary numerical results illustrate the efficiency of the proposed method and demonstrate that it can significantly outperform its non-adaptive counterpart.
format Preprint
id arxiv_https___arxiv_org_abs_2501_11616
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fully Adaptive Zeroth-Order Method for Minimizing Functions with Compressible Gradients
Grapiglia, Geovani Nunes
McKenzie, Daniel
Optimization and Control
We propose an adaptive zeroth-order method for minimizing differentiable functions with $L$-Lipschitz continuous gradients. The method is designed to take advantage of the eventual compressibility of the gradient of the objective function, but it does not require knowledge of the approximate sparsity level $s$ or the Lipschitz constant $L$ of the gradient. We show that the new method performs no more than $O\left(n^{2}ε^{-2}\right)$ function evaluations to find an $ε$-approximate stationary point of an objective function with $n$ variables. Assuming additionally that the gradients of the objective function are compressible, we obtain an improved complexity bound of $O\left(s\log\left(n\right)ε^{-2}\right)$ function evaluations, which holds with high probability. Preliminary numerical results illustrate the efficiency of the proposed method and demonstrate that it can significantly outperform its non-adaptive counterpart.
title Fully Adaptive Zeroth-Order Method for Minimizing Functions with Compressible Gradients
topic Optimization and Control
url https://arxiv.org/abs/2501.11616