A New Training Algorithm for Diagonal Recurrent Neural Network Based on Particle Filter

Author

Xiaolong, Deng ; Pingfang, Xhou

Author_Institution

Dept. of Mech. Eng., Jiangsu Coll. of Inf. Technol., Wuxi, China

Volume

1

fYear

2009

fDate

6-7 June 2009

Firstpage

30

Lastpage

33

Abstract

Based on particle filter, a new training algorithm combining the extended Kalman filter (EKF) for neural network is presented. The new algorithm is firstly applied to train diagonal recurrent neural network (DRNN). A method to evaluate the dynamical performance of DRNN is introduced. Network weights of particles are optimized by the resampling algorithm. Simulation results of the nonlinear dynamical identification verify the validity of the new algorithm.

Keywords

Kalman filters; learning (artificial intelligence); nonlinear filters; recurrent neural nets; diagonal recurrent neural network; extended Kalman filter; network weights; nonlinear dynamical identification; particle filter; resampling algorithm; training algorithm; Computational intelligence; Computer networks; Educational institutions; Information technology; Mechanical engineering; Neural networks; Neurons; Nonlinear dynamical systems; Particle filters; Recurrent neural networks; EKF; diagonal recurrent neural network; nonlinear identification; particle filter;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Natural Computing, 2009. CINC '09. International Conference on

Conference_Location

Wuhan

Print_ISBN

978-0-7695-3645-3

Type

conf

DOI

10.1109/CINC.2009.174

Filename

5231644