Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!howland.reston.ans.net!torn!news!hercules!blam
From: blam@hercules.uwindsor.ca (B Lam)
Subject: More iterations with more training patterns
X-Nntp-Posting-Host: hercules.engn.uwindsor.ca
Message-ID: <D7pLE1.Kts@news.uwindsor.ca>
Sender: news@news.uwindsor.ca (Usenet)
Reply-To: blam@hercules.uwindsor.ca
Organization: Sun Microsystems, Inc.
Date: Thu, 27 Apr 1995 19:42:01 GMT
Lines: 9

Hi! I wrote a training program using BP algorithm. I found out that the network
took more iterations when increase the number of training patterns. I'm wonder
that is there any mathematical explanation of this phenomina? Also when I increase
the number of hidden neurons, the iterations seems to decrease. Is there any NN books
or equations can explain this?

Please send your comments to me blam@engn.uwindsor.ca
BEN

