corner
corner
APS » Journals » Phys. Rev. E » Volume 65 » Issue 5
< Previous Article | Next Article >

Phys. Rev. E 65, 050903(R) (2002) [4 pages]

Learning and predicting time series by neural networks

Download: PDF (66 kB) Buy this article Export: BibTeX or EndNote (RIS)

Ansgar Freking and Wolfgang Kinzel
Institut für Theoretische Physik und Astrophysik, Universität Würzburg, Am Hubland, 97074 Würzburg, Germany

Ido Kanter
Minerva Center and Department of Physics, Bar-Ilan University, Ramat-Gan 52900, Israel

Received 6 December 2001; revised 27 February 2002; published 21 May 2002

Artificial neural networks which are trained on a time series are supposed to achieve two abilities: first, to predict the series many time steps ahead and second, to learn the rule which has produced the series. It is shown that prediction and learning are not necessarily related to each other. Chaotic sequences can be learned but not predicted while quasiperiodic sequences can be well predicted but not learned.

© 2002 The American Physical Society

URL:
http://link.aps.org/doi/10.1103/PhysRevE.65.050903
DOI:
10.1103/PhysRevE.65.050903
PACS:
87.18.Sn, 05.20.-y, 05.45.Tp
 
< Previous Article | Next Article >