Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news3.near.net!paperboy.wellfleet.com!news-feed-1.peachnet.edu!gatech!bloom-beacon.mit.edu!spool.mu.edu!torn!watserv2.uwaterloo.ca!watserv3.uwaterloo.ca!gasser
From: gasser@watnow.uwaterloo.ca (Gasser A. Auda)
Subject: any criteria for choosing supervised networks ?
Message-ID: <D9rHEA.4yr@watserv3.uwaterloo.ca>
Sender: news@watserv3.uwaterloo.ca
Nntp-Posting-Host: watnow.uwaterloo.ca
Organization: University of Waterloo
Date: Tue, 6 Jun 1995 17:18:09 GMT
Lines: 13

Hello:

Did anyone develop a criteria for choosing among different supervised NN's
to solve a classification problem ? In other words, how can the nature of
the supervised classification problem affect the selection of the applied
NN, whether Backprop, RBF, PNN, Projection, Counterprop, Fuzzy ART, .. etc.

Is it trial and error ? or is it just Backprop and that's it ?!

Thanks in advance.

gasser

