Title :
A recursive approach to stochastic optimization via infinitesimal perturbation analysis
Author :
Chong, Edwin K P
Author_Institution :
Sch. of Electr. Eng., Purdue Univ., West Lafayette, IN, USA
Abstract :
We present a general recursive framework for analyzing gradient optimization algorithms driven by infinitesimal perturbation analysis estimates. We give sufficient conditions that guarantee convergence of the algorithm to the optimizing point. We illustrate our results via examples from queueing systems
Keywords :
conjugate gradient methods; discrete event systems; mathematical programming; perturbation techniques; stochastic programming; convergence; general recursive framework; gradient optimization algorithms; infinitesimal perturbation analysis; infinitesimal perturbation analysis estimates; queueing systems; stochastic optimization; Algorithm design and analysis; Costs; Equations; Optimization methods; Performance analysis; Process control; Queueing analysis; Recursive estimation; Stochastic processes; Sufficient conditions;
Conference_Titel :
Decision and Control, 1994., Proceedings of the 33rd IEEE Conference on
Conference_Location :
Lake Buena Vista, FL
Print_ISBN :
0-7803-1968-0
DOI :
10.1109/CDC.1994.411085
