Re: Fuzzy Logic & Neural Nets

Will Dwinnell (predictor@compuserve.com)
Sat, 18 Dec 1999 15:18:43 +0100 (MET)

bv wrote:
"Assuming we do not have prior expert knowledge (or want to avoid any
preconception which would reduce the search space) nor any need to
formalize rules, I believe the comparison then boils down to the fuzzy
engine and classification neural nets.

Rephrased differently. Can I then substitute a fuzzy engine for a
classification net (or a more complex neural net architecture, and
eventually consider separately a rule interpretor and/or other interface
agent around it."

Fuzzy logic is part of a larger fuzzy calculus which may be used any
number of ways. Many people focus on either the fuzzy rule-base model
(fuzzify-fire fuzzy rules-defuzzify) or the even simpler fuzzy truth
value version of ordinary rules bases (almost like certainty factors).
Despite this way of perceiving fuzzy logic, there are many other fuzzy
systems and, as Zadeh himself has noted, anywhere ordinary numbers are
used, fuzzy numbers might be used instead. So, for example, linear
regressions can be transformed into fuzzy linear regressions. There are
any number of fuzzy learning systems, such as fuzzy c-means clustering.

Will Dwinnell
predictor@compuserve.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