Stochastic approximation: Difference between revisions

Revision as of 20:29, 3 November 2008

Revision as of 12:40, 4 November 2008

Line 1:

'''Stochastic approximation''' methods are a family of iterative [[stochastic optimization]] [[algorithm]]s that attempt to find zeroes or extrema of functions which cannot be computed directly, but only estimated via noisy observations. The first, and prototypical, algorithms of this kind were the '''Robbins-Monro''' and '''Kiefer-Wolfowitz''' algorithms.

__NOTOC__

==Robbins-Monro algorithm==

In the [[Herbert Robbins|Robbins]]-Monro algorithm, introduced in 1951<ref name="rm">A Stochastic Approximation Method, Herbert Robbins and Sutton Monro, ''Annals of Mathematical Statistics'' '''22''', #3 (September 1951), pp. 400–407.</ref>, one