An alternative proof for convergence of stochastic approximation algorithms

S. R. Kulkarni; C. S. Horn

doi:10.1109/9.486642

An alternative proof for convergence of stochastic approximation algorithms

S. R. Kulkarni, C. S. Horn

Research output: Contribution to journal › Article › peer-review

9 Scopus citations

Abstract

An alternative proof for convergence of stochastic approximation algorithms is provided. The proof is completely deterministic, very elementary (involving only basic notions of convergence), and direct in that it remains in a discrete setting. An alternative form of the Kushner-Clark condition is introduced and utilized and the results are the first to prove necessity for general gain sequences in a Hilbert space setting.

Original language	English (US)
Pages (from-to)	419-424
Number of pages	6
Journal	IEEE Transactions on Automatic Control
Volume	41
Issue number	3
DOIs	https://doi.org/10.1109/9.486642
State	Published - 1996

All Science Journal Classification (ASJC) codes

Control and Systems Engineering
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/9.486642

Cite this

@article{3f1e0f4a8dc544a884a0bbcde9c2e678,

title = "An alternative proof for convergence of stochastic approximation algorithms",

abstract = "An alternative proof for convergence of stochastic approximation algorithms is provided. The proof is completely deterministic, very elementary (involving only basic notions of convergence), and direct in that it remains in a discrete setting. An alternative form of the Kushner-Clark condition is introduced and utilized and the results are the first to prove necessity for general gain sequences in a Hilbert space setting.",

author = "Kulkarni, {S. R.} and Horn, {C. S.}",

year = "1996",

doi = "10.1109/9.486642",

language = "English (US)",

volume = "41",

pages = "419--424",

journal = "IEEE Transactions on Automatic Control",

issn = "0018-9286",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - An alternative proof for convergence of stochastic approximation algorithms

AU - Kulkarni, S. R.

AU - Horn, C. S.

PY - 1996

Y1 - 1996

N2 - An alternative proof for convergence of stochastic approximation algorithms is provided. The proof is completely deterministic, very elementary (involving only basic notions of convergence), and direct in that it remains in a discrete setting. An alternative form of the Kushner-Clark condition is introduced and utilized and the results are the first to prove necessity for general gain sequences in a Hilbert space setting.

AB - An alternative proof for convergence of stochastic approximation algorithms is provided. The proof is completely deterministic, very elementary (involving only basic notions of convergence), and direct in that it remains in a discrete setting. An alternative form of the Kushner-Clark condition is introduced and utilized and the results are the first to prove necessity for general gain sequences in a Hilbert space setting.

UR - http://www.scopus.com/inward/record.url?scp=0030109229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030109229&partnerID=8YFLogxK

U2 - 10.1109/9.486642

DO - 10.1109/9.486642

M3 - Article

AN - SCOPUS:0030109229

SN - 0018-9286

VL - 41

SP - 419

EP - 424

JO - IEEE Transactions on Automatic Control

JF - IEEE Transactions on Automatic Control

IS - 3

ER -

An alternative proof for convergence of stochastic approximation algorithms

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this