REVIEW 3 cited by
Randomized Derivative-Free Optimization of Noisy Convex Functions
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We propose STARS, a randomized derivative-free algorithm for unconstrained optimization when the function evaluations are contaminated with random noise. STARS takes dynamic, noise-adjusted smoothing step-sizes that minimize the least-squares error between the true directional derivative of a noisy function and its finite difference approximation. We provide a convergence rate analysis of STARS for solving convex problems with additive or multiplicative noise. Experimental results show that (1) STARS exhibits noise-invariant behavior with respect to different levels of stochastic noise; (2) the practical performance of STARS in terms of solution accuracy and convergence rate is significantly better than that indicated by the theoretical result; and (3) STARS outperforms a selection of randomized zero-order methods on both additive and multiplicative-noisy functions.
Forward citations
Cited by 3 Pith papers
-
Robust Airfoil Design Optimization via a Bilevel Model-Based Methodology
A bilevel algorithm combining Bayesian optimization (uncertainty space) and local derivative-free models (design space) finds a RAE2822 airfoil with up to 52% higher lift-to-drag ratio at high Mach conditions, using 2...
-
A Structured Proximal Stochastic Variance Reduced Zeroth-order Algorithm
VR-SZD, a variance-reduced zeroth-order method using orthogonal structured directions, achieves O(d n^{2/3} ε^{-1}) function evaluations for non-convex composite finite-sum problems and linear convergence under the Po...
-
A Structured Tour of Optimization with Finite Differences
A benchmark plus a variance bound showing structured orthogonal direction matrices improve gradient estimation and convergence in finite-difference optimization with little extra generation cost.
Discussion (0). Continue with ORCID to comment.