Randomized-direction stochastic approximation algorithms using deterministic sequences

Author

Xiaoping Xiong ; I-Jeng Wang

Author_Institution

R.H. Smith Sch. of Bus., Maryland Univ., USA

Volume

1

fYear

2002

fDate

8-11 Dec. 2002

Firstpage

285

Abstract

We study the convergence and asymptotic normality of a generalized form of stochastic approximation algorithm with deterministic perturbation sequences. Both one-simulation and two-simulation methods are considered. Assuming a special structure of deterministic sequence, we establish sufficient condition on the noise sequence for AS convergence of the algorithm. Construction of such a special structure of deterministic sequence follows the discussion of asymptotic normality. Finally we discuss ideas on further research in analysis and design of the deterministic perturbation sequences.

Keywords

convergence of numerical methods; randomised algorithms; sequences; simulation; stochastic processes; asymptotic normality; convergence; deterministic perturbation sequences; noise sequence; one-simulation methods; randomized-direction stochastic approximation algorithms; two-simulation methods; Approximation algorithms; Computational modeling; Convergence; Design optimization; Educational institutions; Laboratories; Physics; Stochastic processes; Stochastic resonance; Sufficient conditions;

fLanguage

English

Publisher

ieee

Conference_Titel

Simulation Conference, 2002. Proceedings of the Winter

Print_ISBN

0-7803-7614-5

Type

conf

DOI

10.1109/WSC.2002.1172897

Filename

1172897