Monte Carlo Sampling-Based Methods for Stochastic …

← Back to document page

Monte Carlo Sampling-Based Methods for Stochastic OptimizationTito Homem-de-MelloSchool of BusinessUniversidad Adolfo Iba nezSantiago, uzin BayraksanIntegrated Systems EngineeringThe Ohio State UniversityColumbus, 22, 2014AbstractThis paper surveys the use of Monte Carlo Sampling-Based Methods for Stochastic optimizationproblems. Such Methods are required when as it often happens in practice the model involvesquantities such as expectations and probabilities that cannot be evaluated exactly. While estimationprocedures via sampling are well studied in statistics, the use of such Methods in an optimizationcontext creates new challenges such as ensuring convergence of optimal solutions and optimal val-ues, testing optimality conditions, choosing appropriate sample sizes to balance the effort betweenoptimi

Bene ting from advances in computational power as well as from new theoretical developments, the area of stochastic optimization has evolved considerably in the past few years, with many recent

  Based, Methods, Sampling, Optimization, Stochastic, Sampling based methods for stochastic