Neuronal Architectures for Mixed Estimation and Stabilized Output Error Methods

Kashyap, SK and Raol, JR (2008) Neuronal Architectures for Mixed Estimation and Stabilized Output Error Methods. Journal of System Science and Engineering, System Society of India, 17 (2). pp. 29-36. ISSN 0972-5032

PDF SSI_-_08.pdf - Published Version Restricted to Registered users only Download (3MB)

Abstract

Some architectures based m recurrent neural networks for parameter estimation using least squares mixed estimation and stabilized output error principles are presented. Some error bounds of these neuronal-circuit architectures relative to conventional ones are derived. The results should be useful for implementation of parameter estimation methods for real time applications

Item Type:	Article
Uncontrolled Keywords:	Recurrent neural networks, parameter estimation, stabilized output error method, unstable/augmented systems
Subjects:	MATHEMATICAL AND COMPUTER SCIENCES > Cybernetics, Artificial Intelligence and Robotics
Depositing User:	Dr Sudesh Kumar Kashyap
Date Deposited:	29 Jun 2010 06:07
Last Modified:	29 Jun 2010 06:07
URI:	http://nal-ir.nal.res.in/id/eprint/8454

Actions (login required)

View Item