A wawelet based heuristic to dimension Neural Networks for simple signal approximation

Gabriele Colombini, Davide Sottara, Luca Luccarini, Paola Mello

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

Abstract

Before training a feed forward neural network, one needs to define its architecture. Even in simple feed-forward networks, the number of neurons of the hidden layer is a fundamental parameter, but it is not generally possible to compute its optimal value a priori. It is good practice to start from an initial number of neurons, then build, train and test several different networks with a similar hidden layer size, but this can be excessively expensive when the data to be learned are simple, while some real-time constraints have to be satisfied. This paper shows a heuristic method for dimensioning and initializing a network under such assumptions. The method has been tested on a project for waste water treatment monitoring.

Original languageEnglish (US)
Title of host publicationNeural Nets WIRN09 - Proceedings of the 19th Italian Workshop on Neural Nets
PublisherIOS Press
Pages110-115
Number of pages6
ISBN (Print)9781607500728
DOIs
StatePublished - Jan 1 2009
Externally publishedYes

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume204
ISSN (Print)0922-6389

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint Dive into the research topics of 'A wawelet based heuristic to dimension Neural Networks for simple signal approximation'. Together they form a unique fingerprint.

Cite this