How to optimize discrete-event systems from a single sample path by the score function method

DOI10.1007/BF02055195zbMath0714.93063MaRDI QIDQ751612

Reuven Y. Rubinstein

Publication date: 1990

Published in: Annals of Operations Research (Search for Journal in Brave)

zbMATH Keywords

Monte Carlo optimizationqueueing modelsdiscrete-even systemsoptimization in simulationscore function methodsingle sample path

Mathematics Subject Classification ID

Applications of mathematical programming (90C90) Discrete-time control/observation systems (93C55) Optimal stochastic control (93E20)

Related Items (7)

Stochastic quasigradient methods for optimization of discrete event systems ⋮ Convergence rates for steady-state derivative estimators ⋮ Distributed expert systems for queueing network capacity planning ⋮ Conditioning for variance reduction in estimating the sensitivity of simulations ⋮ Decomposable score function estimators for sensitivity analysis and optimization of queueing networks ⋮ Derivatives of Logarithmic Stationary Distributions for Policy Gradient Reinforcement Learning ⋮ Optimization via simulation: A review

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• On the inclusion of the recombination term in discretizations of the semiconductor device equations
• A stochastic steepest-descent algorithm
• A central-limit-theorem version of \(L=\lambda W\)
• Sensitivity analysis and the ``what if problem in simulation analysis
• Perturbation analysis and optimization of queueing networks
• On the asymptotic distribution of multivariate M-estimates
• Asymptotic properties of statistical estimators in stochastic programming
• Convergence of parameter sensitivity estimates in a stochastic experiment
• Monte Carlo Estimation under Different Distributions Using the Same Simulation
• Sensitivity estimates based on one realization of a stochastic system†
• Strong Convexity Results for Queueing Systems
• Perturbation Analysis Gives Strongly Consistent Sensitivity Estimates for the M/G/1 Queue
• Stochastic convexity and its applications
• Smoothed Functionals and Constrained Stochastic Approximation
• The efficiency and heavy traffic properties of the score function method in sensitivity analysis of queueing models
• Sensitivity Analysis for Simulations via Likelihood Ratios
• Robust Statistics

This page was built for publication: How to optimize discrete-event systems from a single sample path by the score function method