Multiscale Stochastic Approximation for Parametric Optimization of Hidden Markov Models
Shalabh Bhatnagar, Vivek S. Borkar · Probability in the Engineering and Informational Sciences · 1997
A two–time scale stochastic approximation algorithm is proposed for simulation-based parametric optimization of hidden Markov models, as an alternative to the traditional approaches to “infinitesimal perturbation analysis.” Its convergence is analyzed, and a queueing example is presented.