English

A stochastic subspace approach to gradient-free optimization in high dimensions

Optimization and Control 2024-07-08 v2

Abstract

We present a stochastic descent algorithm for unconstrained optimization that is particularly efficient when the objective function is slow to evaluate and gradients are not easily obtained, as in some PDE-constrained optimization and machine learning problems. The algorithm maps the gradient onto a low-dimensional random subspace of dimension \ell at each iteration, similar to coordinate descent but without restricting directional derivatives to be along the axes. Without requiring a full gradient, this mapping can be performed by computing \ell directional derivatives (e.g., via forward-mode automatic differentiation). We give proofs for convergence in expectation under various convexity assumptions as well as probabilistic convergence results under strong-convexity. Our method extends the well-known Gaussian smoothing technique to descent in subspaces of dimension greater than one, opening the doors to new analysis of Gaussian smoothing when more than one directional derivative is used at each iteration. We also provide a finite-dimensional variant of a special case of the Johnson-Lindenstrauss lemma. Experimentally, we show that our method compares favorably to coordinate descent, Gaussian smoothing, gradient descent and BFGS (when gradients are calculated via forward-mode automatic differentiation) on problems from the machine learning and shape optimization literature.

Keywords

Cite

@article{arxiv.2003.02684,
  title  = {A stochastic subspace approach to gradient-free optimization in high dimensions},
  author = {David Kozak and Stephen Becker and Alireza Doostan and Luis Tenorio},
  journal= {arXiv preprint arXiv:2003.02684},
  year   = {2024}
}

Comments

arXiv admin note: substantial text overlap with arXiv:1904.01145

R2 v1 2026-06-23T14:05:11.655Z