Abstract
We present a recurrent MLP1 neural network architecture as a predictive model for time-series caracterization and noise filtering. The approach has been tested on a periodic function corrupted by an additive white noise. Tests have been performed with several noise levels and the method is shown to outperform both linear ARMA2 models and other neural network approximators. Moreover, an improved implementation of the classical OBD3 pruning procedure allows an important reduction of the architecture of the network, without any loss in performance.
Preview
Unable to display preview. Download preview PDF.
References
S.A Billings and C.F Fling: Recurrent radial basis function networks for adaptative noise cancellation, Neural Networks, Vol. 8, No. 2. pp. 273–290, 1995.
I. Cham, S.A Kassam: Interference cancellation using radial basis function networks Signal Processing 47, pp. 247–268, 1995.
T. Czernichow et al.: Short term electrical load forecasting with artificial neural networks, Engineering Intelligent Systems, Vol. 4, No. 2, pp. 85–97, june 1996.
J.T Connor et R.D Martin: Recurrent neural networks and robust time series prediction, IEEE transactions on neural networks, vol. 5, pp. 240–254, march 1994.
S. Garcia-Salicetti: A neural lexical post-processor for improved neural predictive word recognition, ICANN'96.
S. Garcia-Salicetti et al.: From characters to words: dynamical segmentation and predictive neural networks, ICASSP-96.
R. Chentouf: Construction de réseaux de neurones multicouches pour l'approximation, Thèse, 1997, Institut National Polytechnique de Grenoble.
Y. Le Cun, J. Denker et A. Solla: Optimal Brain Damage, NIPS, pp. 598–605, 1990.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lenez, T., Dorizzi, B. (1997). Predictive neural models in noisy environment. In: Gerstner, W., Germond, A., Hasler, M., Nicoud, JD. (eds) Artificial Neural Networks — ICANN'97. ICANN 1997. Lecture Notes in Computer Science, vol 1327. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020290
Download citation
DOI: https://doi.org/10.1007/BFb0020290
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63631-1
Online ISBN: 978-3-540-69620-9
eBook Packages: Springer Book Archive