Randomized Difference Two-Timescale Simultaneous Perturbation Stochastic Approximation Algorithms for Simulation Optimization of Hidden Markov Models

Thumbnail Image
Files
TR_2000-13.pdf(537.72 KB)
No. of downloads: 838
Publication or External Link
Date
2000
Authors
Bhatnagar, Shalabh
Fu, Michael C.
Marcus, Steven I.
Bhatnagar, Shashank
Advisor
Marcus, Steven I.
Fu, Michael C.
Citation
DRUM DOI
Abstract
We proposetwo finite difference two-timescale simultaneous perturbationstochastic approximation (SPSA)algorithmsfor simulation optimization ofhidden Markov models. Stability and convergence of both thealgorithms is proved.<p>Numericalexperiments on a queueing model with high-dimensional parameter vectorsdemonstrate orders of magnitude faster convergence using thesealgorithms over related $(N+1)$-Simulation finite difference analoguesand another two-simulation finite difference algorithm that updates incycles.
Notes
Rights