Edinburgh Research Explorer

Gaussian processes for switching regimes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Related Edinburgh Organisations

Open Access permissions

Open

Documents

http://link.springer.com/chapter/10.1007/978-1-4471-1599-1_94
Original languageEnglish
Title of host publicationICANN 98
Subtitle of host publicationProceedings of the 8th International Conference on Artificial Neural Networks, Skövde, Sweden, 2–4 September 1998
PublisherSpringer London
Pages621-626
Number of pages6
ISBN (Electronic)978-1-4471-1599-1
ISBN (Print)978-3-540-76263-8
DOIs
Publication statusPublished - 1998

Publication series

NamePerspectives in Neural Computing
PublisherSpringer London
ISSN (Print)1431-6854

Abstract

It has been shown that Gaussian processes are a competitive tool for nonparametric regression and classification. Furthermore they are equivalent to neural networks in the limit of an infinite number of neurons. Here we show that the versatility of Gaussian processes at defining different textural characteristics can be used to recognise different regimes in a signal switching between different sources.

Download statistics

No data available

ID: 21891806