On-line EM algorithm and reconstruction of chaotic dynamics

Author

Ishii, Shin ; Sato, Masa-aki

Author_Institution

Nara Inst. of Sci. & Technol., Japan

fYear

1998

fDate

31 Aug-2 Sep 1998

Firstpage

360

Lastpage

369

Abstract

We previously (1998) proposed an online EM algorithm for the normalized Gaussian network model, which is a network of local linear regression units. In this paper, we apply our approach to an identification problem of unknown nonlinear dynamics. Our approach is able to reconstruct the dynamics in shorter learning steps than approaches based on the recurrent neural network model. Even when dynamical variables can partially be observed, our approach is able to well reproduce the trajectory of the observed variables

Keywords

Gaussian processes; chaos; estimation theory; neural nets; nonlinear dynamical systems; optimisation; statistical analysis; chaotic dynamics reconstruction; estimation/maximisation algorithm; local linear regression units; neural network; normalized Gaussian network model; online EM algorithm; unknown nonlinear dynamics; Chaos; Covariance matrix; Electronic mail; Humans; Information processing; Laboratories; Partitioning algorithms; Recurrent neural networks; Stochastic processes; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing VIII, 1998. Proceedings of the 1998 IEEE Signal Processing Society Workshop

Conference_Location

Cambridge

ISSN

1089-3555

Print_ISBN

0-7803-5060-X

Type

conf

DOI

10.1109/NNSP.1998.710666

Filename

710666