Special BISC Seminar: Ch. Borgelt on Possibilistic Graphical Models


Subject: Special BISC Seminar: Ch. Borgelt on Possibilistic Graphical Models
From: Michael Berthold (berthold@icsi.berkeley.edu)
Date: Tue May 02 2000 - 17:55:24 MET DST


*********************************************************************
Berkeley Initiative in Soft Computing (BISC)
*********************************************************************

Special BISC Seminar
320 Soda Hall
Tuesday May 2, 2000
2:00-3:00 pm

-------------------------------------------------------------------
POSSIBILISTIC GRAPHICAL MODELS AND HOW TO LEARN THEM FROM DATA

Christian Borgelt
Otto-von-Guericke-University of Magdeburg
Universitaetsplatz 2, D-39106 Magdeburg, Germany
borgelt@iws.cs.uni-magdeburg.de

Abstract

Graphical models, especially probabilistic ones like Bayesian networks
and Markov-networks, are already well established as powerful tools
for reasoning under uncertainty. The idea underlying them is that
reasoning in multi-dimensional domains, which tends to be infeasible
in the domains as a whole --- and the more so, if uncertainty and/or
imprecision are involved ---, can be made feasible by a decomposition
of the available knowledge. The main advantage of such a decomposition,
which can be described by a graph structure, is that it reduces the
reasoning process to computations in lower-dimensional subspaces.
Although the general scheme is fixed, different types of graphical
models can be distinguished based on the calculus underlying them.
In this talk I focus on possibilistic graphical models --- that is,
graphical models based on possibility theory --- and their semantics,
which differ from the semantics of their better known probabilistic
counterparts. Since a large part of recent research in graphical models
has been devoted to learning, I also discuss the principles of learning
possibilistic graphical models from a database of sample cases.

-------------------------------------------------------------------
Note that Simon Kasif's talk is at the same day, 4pm in Soda 310.

If you ever want to remove yourself from this mailing list,
you can send mail to <Majordomo@EECS.Berkeley.EDUwith 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



This archive was generated by hypermail 2b25 : Tue May 02 2000 - 18:10:27 MET DST