BISC Seminar Announcement, July 8th, 4-5pm, 320 Soda

Frank Hoffmann (fhoffman@cs.berkeley.edu)
Wed, 7 Jul 1999 12:10:44 +0200 (MET DST)

*********************************************************************
Berkeley Initiative in Soft Computing (BISC)
*********************************************************************
************************************************************
B I S C S e m i n a r A n n o u n c e m e n t
************************************************************

Evolving Connectionist Systems and Evolving Fuzzy Neural Networks:
Methods, Tools, Applications


Speaker :
Prof. Nikola (Nik) Kasabov
University of Otago, New Zealand
E-mail: NKasabov@infoscience.otago.ac.nz

Date: Thursday, July 8th, 1999
Time: 4-5pm
Location : 320 Soda Hall

Abstract

The seminar introduces the theory and some applications of one approach to
building intelligent adaptive systems. These systems can learn and improve
continuously over time. They can adapt to new data without forgetting the
previously learned data, can adapt to a changing environment in a real time,
can learn in a "lifelong" learning mode, are able to explain what they have
learned in a linguistically meaningful way. The new approach is called
evolving connectionist systems (ECOS). A new model called evolving fuzzy
neural network (EFuNN) is also introduced. ECOS and EFuNNs learn fast
(possibly in one- pass learning mode) using rules for growing, pruning and
aggregation. Connections and nodes are created, deleted and aggregated
dynamically during the EFuNN operation. Such systems can adapt to new data
in an on-line mode subject to choosing optimal values for a set of
parameters. Both active and passive learning methods have been developed for
ECOS, e.g. sleep-ECO learning. The learning in EFuNNs is based on local
tuning. Applications of ECOS and EFuNNs are demonstrated, such as: adaptive
speech recognition; adaptive intelligent agents on the WWW; adaptive robot
control; adaptive time-series prediction; adaptive intelligent expert
systems. Comparative analysis of EFuNNs and traditional techniques, such as
MLP, SOM, ANFIS, regression analysis,fuzzy C-mean clustering, is performed
on bench mark data sets and real application problems.
Papers and software are available from:
http://divcom.otago.ac.nz/infosci/kel/CBIIS.html

References
Kasabov,N. Foundations of Neural Networks, Fuzzy Systems and Knowledge
Engineering, The MIT Press, CA, MA, 1996.
Amari, S. and Kasabov, N. Brain-like Computing and Intelligent Information
Systems, Springer Verlag, Singapore, 1997.
Kasabov and Kozma (eds) Neuro-fuzzy Techniques for Intelligent Information
Systems, Springer Verlag (Phusica Verlag), Heidelberg, 1999

Bio-data:
Nikola (Nik)Kasabov is Professor of Information Science in the Department of
Information Science, University of Otago, Dunedin, New Zealand. He received
his MSc degree in Computer Science and his PhD degree in Mathematical
Sciences from the technical University in Sofia. Kasabov has published over
200 works in the area of intelligent systems, connectionist and hybrid
connectionist systems, fuzzy systems, expert systems, speech recognition,
and data analysis. He is Director of the research laboratory for Knowledge
Engineering and Computational Intelligence in the Department of Information
Science, University of Otago. Kasabov is the immediate past President of
APNNA - Asia Pacific Neural Network Assembly. He is member of the TC12 group
on Artificial Intelligence of IFIP and also member of the IEEE, INNS, NZCS,
NZRS, ENNS, IEEE Computer Society. He was the general chairman of the First,
the Second and the Third International Conferences on Artificial Neural
Networks and Expert Systems - ANNES'93, ANNES'95 and ANNES'97 (the latter
jointly held with ICONIP'97 and ANZIIS'97) and is the chairman of the
forthcoming ICONIP'99 postconference Workshop on "Future directions for
intelligent systems", Dunedin, 22-24 November 1999.

**********************************************************************
Please direct questions with regard to the contents of the talk
and request for papers to the speaker.
**********************************************************************

-- 
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
If you ever want to remove yourself from this mailing list,
you can send mail to <Majordomo@EECS.Berkeley.EDU> with the following
command in the body of your email message:
unsubscribe bisc-group
or from another account,
unsubscribe bisc-group <your_email_adress>
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Do NOT send unsubscribe requests to bisc-group@cs.berkeley.edu
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

############################################################################ 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