Re: BISC Seminar Announcement, July 16th, 4-5pm, 385 Soda Hall

Anthony Cowden (cowden@sonalysts.com)
Wed, 22 Jul 1998 20:43:09 +0200 (MET DST)

BISCers:

Earl Cox described this methodology in his second book in 1996, and we
currently offer knowledge discovery/data mining services in conjunction
with Earl using his software libraries. In addition, we are in the process
of submitting a proposal to a steel manufacturer for using this technique
to develop a controller for the steel-making process, and we have used this
technique in investigating drug discovery data for a major pharmaceutical
firm.

Tony

At 05:29 PM 7/13/98 -0700, Frank Hoffmann wrote:
>
>************************************************************
>B I S C S e m i n a r A n n o u n c e m e n t
>************************************************************
>
>
>Automatic Construction of Fuzzy Rules. Application to Function
>Approximation and Self-learning Fuzzy Controllers
>
> Ignacio Rojas
> University of Granada, Spain
> Email : irojas@atc.ugr.es
>
>
>Date: Thursday, July 16th, 1998
>Time: 4-5pm
>Location : 385 Soda Hall
>
>
> Abstract
>
>
>Historically, fuzzy rule bases have been constructed by knowledge
>acquisition from
>experts while the weights within neural nets have been learned from
>data. However,
>consulting an expert may be difficult and/or expensive; furthermore,
>translating the
>human operator experience directly into the fuzzy linguistic values
>can be influenced by
>the intuition of the operators and designers, so that the fuzzy control
>rules may be
>incomplete or even contradict each other.
>Two different topics will be discussed. First, we will suppose that
>there exist a set
>of given training examples. A general learning method will be described
>as a framework
>for automatically deriving membership functions and fuzzy if-then rules.
>The identification of the fuzzy system structure and the optimization of the
>parameters defining it are performed in conjunction,
>using a two-phase approach.
>Phase 1 learns the
>membership functions and system rules for a specific structure; Phase 2
>generates a new
>and more suitable topology with the information received from the
>previous step; it
>selects the variables in which the number of membership functions should
>increase and their location.
>A trade-off between the number of membership functions
>and rules and the desired approximation accuracy is also discussed.
>Nevertheless, there are situations in which the knowledge (or a given
>set of training
>data) is not available, it is incomplete or it is inaccurate. Especially
>in the field of control
>application, defining an initial set of fuzzy rules, when no previous
>knowledge is available, constitutes one of the most important problems
>for the design of fuzzy controllers. We will describe the design of
>adaptive and self-learning fuzzy controllers in real
>time, in case only limited prior knowledge of the plant to be controlled
>is available, both in
>terms of the quantity and precision of this information. The algorithm
>does not need a
>mathematical model of the plant, as auxiliary fuzzy systems
>accomplish the adaptation and learning of the parameters of the
>principal fuzzy controller.
>
>
>**********************************************************************
>Please direct questions with regard to the contents of the talk
>and request for papers to the speaker. To unsubscribe send email
>to fhoffman@cs.berkeley.edu NOT !! to bisc-group@diva.eecs.berkeley.edu
>**********************************************************************
>
>
>
>--
>---------------------------------------------------------------------------
>Frank Hoffmann UC Berkeley
>Computer Science Division Department of EECS
>Email: fhoffman@cs.berkeley.edu phone: 1-510-642-8282
>URL: http://http.cs.berkeley.edu/~fhoffman fax: 1-510-642-5775
>---------------------------------------------------------------------------
>
>
*********************************************************************
Anthony Cowden, Manager, Fuzzy Systems Solutions
Sonalysts, Inc.
Fuzzy Systems Solutions: http://www.sonalysts.com/fuzzy.html
Fuzzy Query (TM): http://www.sonalysts.com/fq.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