Minimax efficient finite-difference stochastic gradient estimators using black-box function evaluations

DOI10.1016/j.orl.2020.10.013OpenAlexW3094766279MaRDI QIDQ2661588

Xuhui Zhang, Henry Lam, Hai-Dong Li

Publication date: 7 April 2021

Published in: Operations Research Letters (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2007.04443

zbMATH Keywords

modulus of continuityfinite differenceminimax efficiencyLe Cam's methodstochastic gradient estimationzeroth-order oracle

Mathematics Subject Classification ID

Operations research, mathematical programming (90-XX)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Infinitesimal and finite perturbation analysis for queueing networks
• The score function approach for sensitivity analysis of computer simulation models
• Geometrizing rates of convergence. III
• Statistical estimation and optimal recovery
• Local polynomial regression: Optimal kernels and asymptotic minimax efficiency
• Local linear regression smoothers and their minimax efficiencies
• Random gradient-free minimization of convex functions
• Stochastic simulation: Algorithms and analysis
• Replication Schemes For Limiting Expectations
• Multivariate stochastic approximation using a simultaneous perturbation gradient approximation
• Convergence Properties of Infinitesimal Perturbation Analysis Estimates
• Convergence Rates of Finite-Difference Sensitivity Estimates for Stochastic Systems
• Gradient estimation for discrete-event systems by measure-valued differentiation
• Sensitivity Analysis for Simulations via Likelihood Ratios
• Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic Programming
• On the calculation of definite integrals dependent on a parameter by the monte carlo method
• Introduction to nonparametric estimation