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

被引：0

作者：

Liu, BF

Si, J

机构：

[1] Department of Electrical Engineering, Arizona State University, Tempe

来源：

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS | 1997年 / 44卷 / 11期

关键词：

modeling error bounds; nonlinear dynamic system modeling; recurrent neural networks;

D O I：

10.1109/81.641775

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

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) is an element of R-m.

引用

页码：1086 / 1089

页数：4

共 7 条

[1] ALMEIDA LB, 1987, 1ST P IEEE INT C NEU, P609
[2] GRADIENT METHODS FOR THE OPTIMIZATION OF DYNAMIC-SYSTEMS CONTAINING NEURAL NETWORKS
NARENDRA, KS
PARTHASARATHY, K
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1991, 2 (02): : 252 - 262
[3] Learning State Space Trajectories in Recurrent Neural Networks
Pearlmutter, Barak A.
[J]. NEURAL COMPUTATION, 1989, 1 (02) : 263 - 269
[4] GENERALIZATION OF BACK-PROPAGATION TO RECURRENT NEURAL NETWORKS
PINEDA, FJ
[J]. PHYSICAL REVIEW LETTERS, 1987, 59 (19) : 2229 - 2232
[5] Sontag E, 1990, MATH CONTROL THEORY
[6] BACKPROPAGATION THROUGH TIME - WHAT IT DOES AND HOW TO DO IT
WERBOS, PJ
[J]. PROCEEDINGS OF THE IEEE, 1990, 78 (10) : 1550 - 1560
[7] A Learning Algorithm for Continually Running Fully Recurrent Neural Networks
Williams, Ronald J.
Zipser, David
[J]. NEURAL COMPUTATION, 1989, 1 (02) : 270 - 280

← 1 →