Averaged and decorrelated neural networks as a time-series predictor

U. Naftaly, I. Ginzburg, D. Horn, N. Intrator

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We study the effect of removing temporal structure in the prediction error. We observe that networks which are not optimally trained, exhibit strong temporal structure in their prediction error, which can be eliminated using linear regression. This elimination improves performance significantly, but does not lead to the best performance which is achieved by training networks unitl they do not exhibit any such temporal structure. The improvement in performance of ensemble net averaging does not affect possible temporal sturcture of the error, thus averaging can be performed before or after temporal structure removal. We demonstrate these findings on the sunspot data set.

Original languageEnglish
Title of host publicationProceedings of the 12th IAPR International Conference on Pattern Recognition - Conference B
Subtitle of host publicationPattern Recognition and Neural Networks, ICPR 1994
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages429-432
Number of pages4
ISBN (Electronic)0818662700
StatePublished - 1994
Event12th IAPR International Conference on Pattern Recognition - Conference B: Pattern Recognition and Neural Networks, ICPR 1994 - Jerusalem, Israel
Duration: 9 Oct 199413 Oct 1994

Publication series

NameProceedings - International Conference on Pattern Recognition
Volume2
ISSN (Print)1051-4651

Conference

Conference12th IAPR International Conference on Pattern Recognition - Conference B: Pattern Recognition and Neural Networks, ICPR 1994
Country/TerritoryIsrael
CityJerusalem
Period9/10/9413/10/94

Fingerprint

Dive into the research topics of 'Averaged and decorrelated neural networks as a time-series predictor'. Together they form a unique fingerprint.

Cite this