The application of OBE to neural networks

Author

Jiang, Yan ; He, Qing ; Tong, Tiaosheng ; Dilger, Werner

Author_Institution

Coll. of Electr. & Inf. Eng., Hunan Univ., Changsha, China

Volume

2

fYear

2004

fDate

25-29 July 2004

Firstpage

961

Abstract

In 1989, Singhal and Wu showed that training a feed forward neural network can be viewed as an identification problem for a nonlinear dynamic system using an extended Kalman filter algorithm, which can converge in few iterations. In this paper, a new type of optimal bounding ellipsoid (OBE) algorithm is presented, which is a set-membership identification algorithm based on set theory, and it is shown how it can be used for training feedforward neural networks. The algorithm is compared with backpropagation (BP) and extended Kalman filter (EKF) algorithms and simulation results are presented.

Keywords

Kalman filters; backpropagation; convergence of numerical methods; feedforward neural nets; identification; iterative methods; nonlinear dynamical systems; set theory; BP algorithm; EKF algorithm; backpropagation algorithm; convergence; extended Kalman filter algorithm; feedforward neural networks; iterative methods; neural network training; nonlinear dynamical system; optimal bounding ellipsoid algorithm; set theory; set-membership identification algorithm; Artificial neural networks; Convergence; Educational institutions; Ellipsoids; Feedforward neural networks; Feedforward systems; Neural networks; Noise measurement; Nonlinear equations; Set theory;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on

ISSN

1098-7576

Print_ISBN

0-7803-8359-1

Type

conf

DOI

10.1109/IJCNN.2004.1380063

Filename

1380063