Title :
On the Stochastic Approximation Coefficients
Author_Institution :
Department of Electrical Engineering, Illinois Institute of Technology, Chicago, Ill. 60616.
fDate :
3/1/1974 12:00:00 AM
Abstract :
A simple and straightforward derivation of the optimal form for the Kiefer-Wolfowitz stochastic approximation coefficients is presented. The results follow immediately from the mean-square sense convergence proof for the Kiefer-Wolfowitz algorithm by minimizing the upper bound of the error variance.
Keywords :
Application software; Bridges; Computer applications; Difference equations; Dynamic programming; Functional programming; Markov processes; Mathematical programming; Processor scheduling; Stochastic processes;
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
DOI :
10.1109/TSMC.1974.5409119
