This article is cited in 13 papers

Stochastic Systems

A randomized stochastic approximation algorithm for self-learning

O. N. Granichin, O. A. Izmakova

St. Petersburg State University

Abstract: A new stochastic approximation algorithm with input perturbation for self-learning is designed with test perturbations and has certain useful properties, such as consistency of estimates under almost arbitrary perturbations and preservation of simplicity and performance with the growing size of the state space and increasing number of classes. An example on computer-aided modeling of learning is given to illustrate the performance of the algorithm.

Presented by the member of Editorial Board: A. V. Nazin

Received: 27.05.2004

 English version: Automation and Remote Control, 2005, 66:8, 1239–1248

Bibliographic databases:

• 
• 
•