Re: Fuzzy Logic & Neural Nets

vladik (vladik@cs.utep.edu)
Wed, 8 Dec 1999 17:03:25 +0100 (MET)

One of the main applications of fuzzy logic is to formalize expert knowledge
formulate in terms of words from natural language. For example, an expert
controller can express his control strategy by using rules like "if the
obstacle is reasonably close brake harshly". The computerized system cannot use
this information because the expert did not (and usually cannot) exactly
specify what he means by "reasonably close" (5 meters? 20?), and what he means
by harshly. This is where fuzzy logic methodology helps: it enables us to
translate the expert rules into a precise control strategy. For that, we
interview an expert, get membership function from him, and use formulas of
fuzzy control to get a first approximation. After that, we can fine-tune the
resulting strategy.

If expert rules is the only information we have, then fuzzy logic is the only
tool we can use, neural networks cannot help. For a NN to work, we need samples
of control.

In real life, we often have BOTH expert rules and records of actual control, we
the best way is to use BOTH methods (fuzzy and neural). In this sense, these
two formalisms are not competitors but complement each other. Suffice is to say
that after going thorugh a pure fuzzy logic stage, most Japanese appliances
moved to neuro-fuzzy.

Vladik

Vladik Kreinovich
Department of Computer Science
University of Texas at El Paso
El Paso, TX 79968

> Date: Wed, 8 Dec 1999 11:02:18 +0100 (MET)
> Originator: fuzzy-mail@dbai.tuwien.ac.at
> From: "bv" <info@foretrade.com>
> To: nafips-l@sphinx.Gsu.EDU
> Subject: Fuzzy Logic & Neural Nets
> X-Listprocessor-Version: 8.2.07 -- ListProc(tm) by CREN
> X-Comment: Fuzzy Distribution List
> X-Sender: fuzzy-mail@dbai.tuwien.ac.at
>
> I hope this question will not offend anyone.
>
> Is there anything in fuzzy logic neural nets can't already handle ?
> RBF, Probabilistic, Kohonen nets and the likes already handle fuzzy
> classification quite neatly, don;'t they?
>
> Clarifications most welcome.
>
> Regards
>
> Bruno VOISIN
> voisin.bruno@wanadoo.fr
>
>
>
>
> ############################################################################
> 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 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