Error estimation of recurrent neural network models trained on a finite set of initial values

Binfan Liu, Jennie Si

Research output: Contribution to journalArticle

Abstract

This letter addresses the problem of estimating training error bounds of state and output trajectories for a class of recurrent neural networks as models of nonlinear dynamic systems. The bounds are obtained provided that the models have been trained on N trajectories with N independent random initial values which are uniformly distributed over [a,b]m ∈ Rm.

Original languageEnglish (US)
Pages (from-to)1086-1089
Number of pages4
JournalIEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications
Volume44
Issue number11
DOIs
StatePublished - 1997

Fingerprint

Recurrent neural networks
Error analysis
Trajectories
Dynamical systems

Keywords

  • Modeling error bounds
  • Nonlinear dynamic system modeling
  • Recurrent neural networks

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

@article{521e675599304d9281c03100c8f3fe87,
title = "Error estimation of recurrent neural network models trained on a finite set of initial values",
abstract = "This letter addresses the problem of estimating training error bounds of state and output trajectories for a class of recurrent neural networks as models of nonlinear dynamic systems. The bounds are obtained provided that the models have been trained on N trajectories with N independent random initial values which are uniformly distributed over [a,b]m ∈ Rm.",
keywords = "Modeling error bounds, Nonlinear dynamic system modeling, Recurrent neural networks",
author = "Binfan Liu and Jennie Si",
year = "1997",
doi = "10.1109/81.641775",
language = "English (US)",
volume = "44",
pages = "1086--1089",
journal = "IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications",
issn = "1549-8328",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "11",

}

TY - JOUR

T1 - Error estimation of recurrent neural network models trained on a finite set of initial values

AU - Liu, Binfan

AU - Si, Jennie

PY - 1997

Y1 - 1997

N2 - This letter addresses the problem of estimating training error bounds of state and output trajectories for a class of recurrent neural networks as models of nonlinear dynamic systems. The bounds are obtained provided that the models have been trained on N trajectories with N independent random initial values which are uniformly distributed over [a,b]m ∈ Rm.

AB - This letter addresses the problem of estimating training error bounds of state and output trajectories for a class of recurrent neural networks as models of nonlinear dynamic systems. The bounds are obtained provided that the models have been trained on N trajectories with N independent random initial values which are uniformly distributed over [a,b]m ∈ Rm.

KW - Modeling error bounds

KW - Nonlinear dynamic system modeling

KW - Recurrent neural networks

UR - http://www.scopus.com/inward/record.url?scp=0031271276&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031271276&partnerID=8YFLogxK

U2 - 10.1109/81.641775

DO - 10.1109/81.641775

M3 - Article

VL - 44

SP - 1086

EP - 1089

JO - IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications

JF - IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications

SN - 1549-8328

IS - 11

ER -