Time Series Generation by Recurrent Neural Networks

Authors:
A. Priel;I. Kanter
Affiliations:
Department of Physics and Minerva Center, Bar-Ilan University, 52900 Ramat-Gan, Israel, Center for Brain Sciences, Bar-Ilan University, Ramat-Gan, Israel e-mail: priel@mail.biu.ac.ilido@kanter1.ph.biu.ac.il
Venue:
Annals of Mathematics and Artificial Intelligence
Year:
2003

Citing 5
Cited 0

Modular construction of time-delay neural networks for speech recognition

Neural Computation
Introduction to the theory of neural computation

Introduction to the theory of neural computation
Analytical study of the interplay between architecture and predictability

NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
Handbook of Mathematical Functions, With Formulas, Graphs, and Mathematical Tables,

Handbook of Mathematical Functions, With Formulas, Graphs, and Mathematical Tables,
Computational capabilities of recurrent NARX neural networks

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

Quantified Score

Hi-index	0.00

Visualization

Abstract

The properties of time series, generated by continuous valued feed-forward networks in which the next input vector is determined from past output values, are studied. Asymptotic solutions developed suggest that the typical stable behavior is (quasi) periodic with attractor dimension that is limited by the number of hidden units, independent of the details of the weights. The results are robust under additive noise, except for expected noise-induced effects – attractor broadening and loss of phase coherence at large times. These effects, however, are moderated by the size of the network N.