System identification using neural network with parametric sigmoid function

M. Hasheminejad, Junichi Murata, K. Hirasawa

Research output: Contribution to journalArticle

Abstract

Nonlinear systems can be modeled by neural networks. However choice of suitable network architecture is the most important problem. And 'how to find the best activation function' is a persistent aspect of the architecture design. Here we have proposed a sigmoid function with one parameter which provides us not only the reduction of error bound but also the opportunity of obtaining better insight into the systems.

Original languageEnglish
Pages (from-to)39-44
Number of pages6
JournalArtificial Neural Networks in Engineering - Proceedings (ANNIE'94)
Volume4
Publication statusPublished - 1994

Fingerprint

Identification (control systems)
Neural networks
Network architecture
Nonlinear systems
Chemical activation

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

System identification using neural network with parametric sigmoid function. / Hasheminejad, M.; Murata, Junichi; Hirasawa, K.

In: Artificial Neural Networks in Engineering - Proceedings (ANNIE'94), Vol. 4, 1994, p. 39-44.

Research output: Contribution to journalArticle

@article{fe41f72510df48029d6d4c7b41ef417f,
title = "System identification using neural network with parametric sigmoid function",
abstract = "Nonlinear systems can be modeled by neural networks. However choice of suitable network architecture is the most important problem. And 'how to find the best activation function' is a persistent aspect of the architecture design. Here we have proposed a sigmoid function with one parameter which provides us not only the reduction of error bound but also the opportunity of obtaining better insight into the systems.",
author = "M. Hasheminejad and Junichi Murata and K. Hirasawa",
year = "1994",
language = "English",
volume = "4",
pages = "39--44",
journal = "Artificial Neural Networks in Engineering - Proceedings (ANNIE'94)",

}

TY - JOUR

T1 - System identification using neural network with parametric sigmoid function

AU - Hasheminejad, M.

AU - Murata, Junichi

AU - Hirasawa, K.

PY - 1994

Y1 - 1994

N2 - Nonlinear systems can be modeled by neural networks. However choice of suitable network architecture is the most important problem. And 'how to find the best activation function' is a persistent aspect of the architecture design. Here we have proposed a sigmoid function with one parameter which provides us not only the reduction of error bound but also the opportunity of obtaining better insight into the systems.

AB - Nonlinear systems can be modeled by neural networks. However choice of suitable network architecture is the most important problem. And 'how to find the best activation function' is a persistent aspect of the architecture design. Here we have proposed a sigmoid function with one parameter which provides us not only the reduction of error bound but also the opportunity of obtaining better insight into the systems.

UR - http://www.scopus.com/inward/record.url?scp=0028740890&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028740890&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0028740890

VL - 4

SP - 39

EP - 44

JO - Artificial Neural Networks in Engineering - Proceedings (ANNIE'94)

JF - Artificial Neural Networks in Engineering - Proceedings (ANNIE'94)

ER -