Re: Please help in Neural Network Implementaion !!


Subject: Re: Please help in Neural Network Implementaion !!
From: Myriam Othmane (myriam.othmane@wanadoo.be)
Date: Mon Feb 07 2000 - 22:01:12 MET


Hello,
For non-linear problem all NN are 'universal approximator' 1, the MLP
(multipli layer perceptron) is much easier to implement and you can find
programs ready to use (EM SAS, NN module from SPSS). The RBF (Radial Basis
Function) is for me more insensitif to outliers.
With the MLP and the RBF you can predict the order of Y. Your algorithm will
be able to fit your news X in the range [..., Order_Ymin,..., Order_Ymax,
....].
With these method you must take care to have enough data (row) to split your
dataset in two folds : training set and the validate set (and may-be a test
set to estimate accurately the error of your model).

regards,
Manuel.

1 : depends on the design of your network (number of hidden neuron and
links).

Joe Smith <jsmith@currentreviews.com> a écrit dans le message :
3899DAA8.694AA333@currentreviews.com...
> Dear Chin:
>
> Chin Wei Chuen wrote:
> >
> > Regarding the prediction of Y vs [X], which in fact is a pattern
recognition
> > problem, there are two classes of Neural Network algorithms,i.e.
supervised and
> > unsupervised training methods.
>
> Which of the two classes do you think could be more accurate in resolving
the
> problem ?
>
> > In your case of unsupervised training (that is: you don't have the
actual Y
> > values) then you would need to use NN algorithms like Kohonen or
Hopfield
> > networks. Feel free to read up on those
>
> If I understand you correctly, using a Hopfield network design "I do not
need
> the real values of Y" just the rank order of Y. Is that correct ?
>
> > For supervised training, which I am more familiar with, you would
require some
> > training data that includes both X and Y values to form the weights and
later
> > feeding your newly formed subjects into the network to predict the
outcome.
> > Traditional supervised Neural Network algorithms include the Multi-Layer
> > Perceptron (MLP) network using backprop or the Radial Basis Function.
>
> Out of all the above network designs, which one is best at solving
non-linear
> problems ?
>
> Thanks again for the help ...
>
> Joe Smith
> jsmith@currentreviews.com

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



This archive was generated by hypermail 2b25 : Thu Apr 06 2000 - 15:59:42 MET DST