Recursive least squares approach to learning in recurrent neural networks

Author

Parisi, R. ; Claudio, E. D Di ; Rapagnetta, A. ; Orlandi, G.

Author_Institution

INFOCOM Dept., Rome Univ., Italy

Volume

2

fYear

1996

fDate

3-6 Jun 1996

Firstpage

1350

Abstract

This paper presents a new approach to learning in recurrent neural networks, based on the descent of the error functional in the space of the linear outputs of the neurons (neuron space approach). At each step of the learning process a linear system is solved for the weights using a recursive least squares technique. This approach, with respect to traditional gradient-based algorithms, guarantees better performances from the point of view of both the speed of convergence and the numerical robustness

Keywords

convergence of numerical methods; learning (artificial intelligence); least squares approximations; recurrent neural nets; convergence; error functional descent; gradient-based algorithms; learning; neuron space approach; numerical robustness; recurrent neural networks; recursive least-squares technique; Adaptive filters; Convergence of numerical methods; Intelligent networks; Least squares methods; Linear systems; Neural networks; Neurons; Nonlinear dynamical systems; Recurrent neural networks; Robustness;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1996., IEEE International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-7803-3210-5

Type

conf

DOI

10.1109/ICNN.1996.549095

Filename

549095