A Global Optimization Approach to Neural Network Training

dc.contributor.author	Lagaris, I. E.	en
dc.contributor.author	Voglis, C	en
dc.date.accessioned	2015-11-24T17:01:19Z
dc.date.available	2015-11-24T17:01:19Z
dc.identifier.uri	https://olympias.lib.uoi.gr/jspui/handle/123456789/10907
dc.rights	Default Licence	-
dc.subject	Arti?cial neural networks,	en
dc.subject	global optimization,	en
dc.subject	multistart.	en
dc.title	A Global Optimization Approach to Neural Network Training	en
heal.abstract	We study effective approaches for training arti?cial neural networks (ANN). We argue that local optimization methods by themselves are not suited for that task. In fact we show that global optimization methods are absolutely necessary if the training is required to be robust. This is so because the objective function under consideration possesses a multitude of minima while only a few may correspond to acceptable solutions that generalize well.	en
heal.access	campus	-
heal.fullTextAvailability	TRUE	-
heal.journalName	Neural Parallel and Scientific Computations	en
heal.journalType	peer reviewed	-
heal.language	en	-
heal.publicationDate	2006	-
heal.recordProvider	Πανεπιστήμιο Ιωαννίνων. Σχολή Θετικών Επιστημών. Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικής	el
heal.type	journalArticle	-
heal.type.el	Άρθρο Περιοδικού	el
heal.type.en	Journal article	en

Αρχεία

Προβολή: 1 - 1 of 1