Reconstructions and predictions of nonlinear dynamical systems by Rao-Blackwellised sequential Monte Carlo

Author

Soma, T. ; Yosui, K. ; Matsumoto, Tad

Author_Institution

Dept. of Electr., Electron. & Comput. Eng., Waseda Univ., Tokyo, Japan

Volume

2

fYear

2003

fDate

6-10 April 2003

Abstract

Sequential Monte Carlo (SMC) is a powerful sampling based inference/learning algorithm for Bayesian scheme. The purpose of this paper is two fold. It first attempts to reconstruct and predict nonlinear dynamical systems from one dimensional data which arrives in a sequential manner instead of batch manner. Second purpose is to test the performance of the Rao-Blackwellisation in reconstructing and predicting nonlinear dynamical systems. We demonstrate that Rao-Blackwellised sequential Monte Carlo (RBSMC) on a chaotic time series prediction problem outperforms generic SMC.

Keywords

Monte Carlo methods; inference mechanisms; learning (artificial intelligence); prediction theory; signal reconstruction; signal sampling; time series; Bayesian scheme; Rao-Blackwellised sequential Monte Carlo; chaotic time series prediction; nonlinear dynamical system prediction; nonlinear dynamical system reconstruction; one dimensional data; sampling based inference/learning algorithm; sequential Monte Carlo; Bayesian methods; Chaos; Equations; Monte Carlo methods; Nonlinear dynamical systems; Power engineering and energy; Power engineering computing; Sampling methods; Sliding mode control; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1202482

Filename

1202482