13
0

Variance-Reduced Gradient Estimator for Nonconvex Zeroth-Order Distributed Optimization

Abstract

This paper investigates distributed zeroth-order optimization for smooth nonconvex problems. We propose a novel variance-reduced gradient estimator, which randomly renovates one orthogonal direction of the true gradient in each iteration while leveraging historical snapshots for variance correction. By integrating this estimator with gradient tracking mechanism, we address the trade-off between convergence rate and sampling cost per zeroth-order gradient estimation that exists in current zeroth-order distributed optimization algorithms, which rely on either the 2-point or 2d2d-point gradient estimators. We derive a convergence rate of O(d52/m)\mathcal{O}(d^{\frac{5}{2}}/m) for smooth nonconvex functions in terms of sampling number mm and problem dimension dd. Numerical simulations comparing our algorithm with existing methods confirm the effectiveness and efficiency of the proposed gradient estimator.

View on arXiv
Comments on this paper