-------------------------------------------------------------------
Speaker:        Prof. Norman Foo
                Knowledge Systems Group
                Artificial Intelligence Laboratory
                School of Computer Science and Engineering
                The University of New South Wales
                Australia

Topic:          "Coherence of Theories"

Date:           Monday, 21 Oct 2002

Time:           4:00 p.m. - 5:00 p.m.

Venue:          Lecture Theater F
                (Leung Yat Sing Lecture Theater, via lift nos. 25/26)
                HKUST


Abstract:

The vocabulary of scientific theories can often be divided into two parts
-- the observational and the theoretical.  The observational vocabulary
comprises terms that refer to experimentally measurable entities, while
the theoretical comprises the "hidden" variables. I will cite examples
from various disciplines to illustrate these divisions.  Given two
theories with different theoretical content but identical observational
predictions, are there rational ways to choose between them?  One
recurrent theme in the philosophy of science is that of coherence.
Informally, a highly coherent theory has the following characteristics: it
binds tightly together; it provides succinct explanations; it is
irredundant.  The relevance of coherence to the problem of theory choice
is that it has been proposed as a major criterion for such choice.  This
talk is about one way to quantify the hitherto informal notion of
coherence.  The approach is to initially use classical logic together with
AI ideas of abduction and simple combinatorics to measure theory
coherence.  At the end I will suggest how to liberate the main ideas of
the talk from logic.



***********************
Biography:

Norman Foo graduated B.E. (1965) and ME (1966) in Electrical Engineering,
Canterbury University, and M.A. (1970) and PhD (1974) in Computer and
Communication Sciences, University of Michigan.  He has been assistant and
visiting associate professor in SUNY Binghamton and visiting professor in
the IBM Systems Research Institute and T.J. Watson Research Laboratories
in Hawthorne, New York, and in the University of Birmingham, England. From
1975 he was with the Basser Department of Computer, Sydney University
where he eventually held a personal chair as Professor of Knowledge
Systems.  In 1996 he moved to the Department of Artificial Intelligence,
School of Computer Science and Engineering, University of New South Wales
where he is Professor of Computer Science and Engineering.

Norman's research has ranged in the past from algorithm analysis, abstract
datatypes, complexity theory, and modelling and simulation to his present
interests in logic programming, knowledge representation, artificial
intelligence logics, and cognitive science. He has graduated 17 doctoral
and 3 masters students.  He is on the editorial board of the International
Journal of General Systems, and the Journal of Knowledge and Information
Systems.  In 1998 the Australian Research Council gave him a five-year
Special Investigator Award.  From 2003 he will lead the Knowledge
Representation and Reasoning Program in Australia's new National ICT
Australia (NICTA) research institute based in UNSW and the ANU.


                For enquiry, please call 2358 7008
                        ** All are Welcome **
---------------------------------------------------------------------------