Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks

Kotaro Hirasawa, Takayuki Furuzuki, Junichi Murata, ChunZhi Jin, Kazuaki Yotsumoto, Hironobu Katagiri

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, a method for identifying nonlinear dynamic systems with noise is proposed by using Probabilistic Universal Learning Networks (PrULNs). PrULNs are extensions of Universal Learning Networks (ULNs). ULNs form a superset of neural networks and were proposed to provide a universal framework for modeling and control of nonlinear large-scale complex systems. But the ULN does not provide any stochastic characteristics of the signals propagating through it. The PrULNs are equipped with machinery to calculate stochastic properties of signals and to train network parameters so that the signals behave with the pre-specified stochastic properties. On the other hand, it is generally recognized that there exists an overfitting problem when identification of nonlinear dynamic systems with noise is done by neural networks. In this paper, it is shown from simulation results of identification of a nonlinear robot dynamics that PrULNs are useful for avoiding the overfitting.

Original languageEnglish
Title of host publicationProceedings of the International Joint Conference on Neural Networks
Place of PublicationUnited States
PublisherIEEE
Pages2123-2128
Number of pages6
Volume3
Publication statusPublished - 1999
Externally publishedYes
EventInternational Joint Conference on Neural Networks (IJCNN'99) - Washington, DC, USA
Duration: 1999 Jul 101999 Jul 16

Other

OtherInternational Joint Conference on Neural Networks (IJCNN'99)
CityWashington, DC, USA
Period99/7/1099/7/16

Fingerprint

Dynamical systems
Neural networks
Machinery
Large scale systems
Robots

ASJC Scopus subject areas

  • Software

Cite this

Hirasawa, K., Furuzuki, T., Murata, J., Jin, C., Yotsumoto, K., & Katagiri, H. (1999). Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks. In Proceedings of the International Joint Conference on Neural Networks (Vol. 3, pp. 2123-2128). United States: IEEE.

Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks. / Hirasawa, Kotaro; Furuzuki, Takayuki; Murata, Junichi; Jin, ChunZhi; Yotsumoto, Kazuaki; Katagiri, Hironobu.

Proceedings of the International Joint Conference on Neural Networks. Vol. 3 United States : IEEE, 1999. p. 2123-2128.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hirasawa, K, Furuzuki, T, Murata, J, Jin, C, Yotsumoto, K & Katagiri, H 1999, Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks. in Proceedings of the International Joint Conference on Neural Networks. vol. 3, IEEE, United States, pp. 2123-2128, International Joint Conference on Neural Networks (IJCNN'99), Washington, DC, USA, 99/7/10.
Hirasawa K, Furuzuki T, Murata J, Jin C, Yotsumoto K, Katagiri H. Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks. In Proceedings of the International Joint Conference on Neural Networks. Vol. 3. United States: IEEE. 1999. p. 2123-2128
Hirasawa, Kotaro ; Furuzuki, Takayuki ; Murata, Junichi ; Jin, ChunZhi ; Yotsumoto, Kazuaki ; Katagiri, Hironobu. / Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks. Proceedings of the International Joint Conference on Neural Networks. Vol. 3 United States : IEEE, 1999. pp. 2123-2128
@inproceedings{1fff9cbde48b44eaabc2b13cf524f805,
title = "Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks",
abstract = "In this paper, a method for identifying nonlinear dynamic systems with noise is proposed by using Probabilistic Universal Learning Networks (PrULNs). PrULNs are extensions of Universal Learning Networks (ULNs). ULNs form a superset of neural networks and were proposed to provide a universal framework for modeling and control of nonlinear large-scale complex systems. But the ULN does not provide any stochastic characteristics of the signals propagating through it. The PrULNs are equipped with machinery to calculate stochastic properties of signals and to train network parameters so that the signals behave with the pre-specified stochastic properties. On the other hand, it is generally recognized that there exists an overfitting problem when identification of nonlinear dynamic systems with noise is done by neural networks. In this paper, it is shown from simulation results of identification of a nonlinear robot dynamics that PrULNs are useful for avoiding the overfitting.",
author = "Kotaro Hirasawa and Takayuki Furuzuki and Junichi Murata and ChunZhi Jin and Kazuaki Yotsumoto and Hironobu Katagiri",
year = "1999",
language = "English",
volume = "3",
pages = "2123--2128",
booktitle = "Proceedings of the International Joint Conference on Neural Networks",
publisher = "IEEE",

}

TY - GEN

T1 - Identification of nonlinear dynamic systems by using Probabilistic Universal Learning Networks

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

AU - Murata, Junichi

AU - Jin, ChunZhi

AU - Yotsumoto, Kazuaki

AU - Katagiri, Hironobu

PY - 1999

Y1 - 1999

N2 - In this paper, a method for identifying nonlinear dynamic systems with noise is proposed by using Probabilistic Universal Learning Networks (PrULNs). PrULNs are extensions of Universal Learning Networks (ULNs). ULNs form a superset of neural networks and were proposed to provide a universal framework for modeling and control of nonlinear large-scale complex systems. But the ULN does not provide any stochastic characteristics of the signals propagating through it. The PrULNs are equipped with machinery to calculate stochastic properties of signals and to train network parameters so that the signals behave with the pre-specified stochastic properties. On the other hand, it is generally recognized that there exists an overfitting problem when identification of nonlinear dynamic systems with noise is done by neural networks. In this paper, it is shown from simulation results of identification of a nonlinear robot dynamics that PrULNs are useful for avoiding the overfitting.

AB - In this paper, a method for identifying nonlinear dynamic systems with noise is proposed by using Probabilistic Universal Learning Networks (PrULNs). PrULNs are extensions of Universal Learning Networks (ULNs). ULNs form a superset of neural networks and were proposed to provide a universal framework for modeling and control of nonlinear large-scale complex systems. But the ULN does not provide any stochastic characteristics of the signals propagating through it. The PrULNs are equipped with machinery to calculate stochastic properties of signals and to train network parameters so that the signals behave with the pre-specified stochastic properties. On the other hand, it is generally recognized that there exists an overfitting problem when identification of nonlinear dynamic systems with noise is done by neural networks. In this paper, it is shown from simulation results of identification of a nonlinear robot dynamics that PrULNs are useful for avoiding the overfitting.

UR - http://www.scopus.com/inward/record.url?scp=0033348747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033348747&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033348747

VL - 3

SP - 2123

EP - 2128

BT - Proceedings of the International Joint Conference on Neural Networks

PB - IEEE

CY - United States

ER -