On predicting the optimal number of hidden nodes

Alan Thomas, Miltiadis Petridis, Simon Walters, Mohammad Malekshahi Gheytassi, Robert Morgan

Research output: Chapter in Book/Conference proceeding with ISSN or ISBNConference contribution with ISSN or ISBN

Abstract

Abstract- Determining the optimal number of hidden nodes isthe most challenging aspect of Artificial Neural Network(ANN) design. To date, there are still no reliable methods ofdetermining this a priori, as it depends on so many domainspecificfactors. Current methods which take these intoaccount, such as exhaustive search, growing and pruning andevolutionary algorithms are not only inexact, but alsoextremely time consuming - in some cases prohibitively so. Anovel approach embodied in a system called Heurix isintroduced. This rapidly predicts the optimal number ofhidden nodes from a small number of sample topologies. It canbe configured to favour speed (low complexity), accuracy, or abalance between the two. Single hidden layer feedforwardnetworks (SLFNs) can be built twenty times faster, and with ageneralisation error of as little as 0.4% greater than thosefound by exhaustive search.
Original languageEnglish
Title of host publication2015 international conference on computational science and computational intelligence
Place of PublicationLas Vegas
PublisherCPS/IEEE
Pages565-570
Number of pages6
ISBN (Print)9781467397957
DOIs
Publication statusPublished - 8 Dec 2015
Event2015 international conference on computational science and computational intelligence - Las Vegas, 7-9 December 2015
Duration: 8 Dec 2015 → …

Conference

Conference2015 international conference on computational science and computational intelligence
Period8/12/15 → …

Fingerprint Dive into the research topics of 'On predicting the optimal number of hidden nodes'. Together they form a unique fingerprint.

  • Cite this

    Thomas, A., Petridis, M., Walters, S., Malekshahi Gheytassi, M., & Morgan, R. (2015). On predicting the optimal number of hidden nodes. In 2015 international conference on computational science and computational intelligence (pp. 565-570). Las Vegas: CPS/IEEE. https://doi.org/10.1109/CSCI.2015.33