Please use this identifier to cite or link to this item:
Type: Artigo de evento
Title: Second-order Training For Recurrent Neural Networks Without Teacher-forcing
Author: Von Zuben Fernando J.
de Andrade Netto Marcio L.
Abstract: Neural networks with external recurrences can be successfully applied to nonlinear autoregressive moving average modeling. The process of weight adjustment is presented as a nonlinear optimization problem in the N-dimensional Euclidean space, where N is the number of adjustable weights. The least-squares criterion can be effectively minimized using a version of the conjugate gradient algorithm. Expending about the same amount of computation necessary to obtain the gradient, the required second-order information is calculated exactly. A simulation example confirms the efficacy of the training process when applied to time series prediction. Contrary to the proposed method, teacher-forced learning is shown to be ill-suited for multi-step prediction.
Editor: IEEE, Piscataway, NJ, United States
Rights: fechado
Identifier DOI: 
Date Issue: 1995
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
2-s2.0-0029511875.pdf456.74 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.