English

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

Optimization and Control 2025-07-16 v3

Abstract

We propose an adaptive zeroth-order method for minimizing differentiable functions with LL-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 ss or the Lipschitz constant LL of the gradient. We show that the new method performs no more than O(n2ϵ2)O\left(n^{2}\epsilon^{-2}\right) function evaluations to find an ϵ\epsilon-approximate stationary point of an objective function with nn variables. Assuming additionally that the gradients of the objective function are compressible, we obtain an improved complexity bound of O(slog(n)ϵ2)O\left(s\log\left(n\right)\epsilon^{-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.

Keywords

Cite

@article{arxiv.2501.11616,
  title  = {Fully Adaptive Zeroth-Order Method for Minimizing Functions with Compressible Gradients},
  author = {Geovani Nunes Grapiglia and Daniel McKenzie},
  journal= {arXiv preprint arXiv:2501.11616},
  year   = {2025}
}

Comments

V3: Added two new benchmark algorithms and a new sparse gradient benchmarking function

R2 v1 2026-06-28T21:11:32.891Z