Application of stochastic approximation techniques in neural modelling and control
| Autor: | C. Renotte, A. Vande Wouwer, M. Remy |
|---|---|
| Rok vydání: | 2003 |
| Předmět: |
Identification (information)
Simultaneous perturbation stochastic approximation Mathematical optimization Artificial neural network Control and Systems Engineering Computer science PID controller Minification Stochastic neural network Stochastic approximation Smoothing Computer Science Applications Theoretical Computer Science |
| Zdroj: | International Journal of Systems Science. 34:851-863 |
| ISSN: | 1464-5319 0020-7721 |
| DOI: | 10.1080/00207720310001640296 |
| Popis: | Learning, i.e. estimation of weights and biases in neural networks, involves the minimization of an output error criterion, a problem which is usually solved using back-propagation algorithms. This paper aims to assess the potential of simultaneous perturbation stochastic approximation (SPSA) algorithms to handle this minimization problem. In particular, a variation of the first-order SPSA algorithm that makes use of several numerical artifices including adaptive gain sequences, gradient smoothing and a step rejection procedure is developed. For illustration purposes, several application examples in the identification and control of nonlinear dynamic systems are presented. This numerical evaluation includes the development of neural network models as well as the design of a model-based predictive neural PID controller. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|