anand@yarra-glen.aaii.oz.au (Anand Rao) (04/18/91)
Call for Papers
---------------
IJCAI-91 WORKSHOP ON THEORETICAL AND PRACTICAL DESIGN OF RATIONAL AGENTS
Sunday, August 25, 1991
Workshop Description
--------------------
During the last few years, AI researchers have become increasingly
concerned with the problem of designing architectures for intelligent
systems that are required to operate in real-world environments. Based
on the sensory input from their environment, these systems must be able
to choose and execute appropriate courses of action to further their (or
their designer's) goals. Moreover, this must be done under possibly
stringent constraints on both time and information.
There has emerged a range of approaches to the design of such systems.
At one extreme are situated automata and related mechanisms, which
incorporate compiled or ``hard-wired'' capabilities to perform their
tasks. At the other extreme are situated planners or rational agents,
which perform much of their reasoning and planning in real time as
they interact with their environment.
This workshop will focus on the rational-agency approach to the design
of AI systems. Its purpose is to bring together researchers working
on various aspects of rational agency and on the design of systems
based on this approach. In the workshop, we will consider the
philosophical foundations and logical formalizations of rational
agency, the role of decision theory in deliberation and meta-level
reasoning, and the design and performance evaluation of such rational
agents in experimental and real-world domains.
Topics of Interest
------------------
Formalizations of Rational Agency
-- logics of beliefs, desires, and intentions
-- modelling quantitative aspects like probability, payoff, and utility
-- theories of belief, desire, and intention revision
Deliberation
-- role of decision-theoretic techniques
-- integration of decision-theoretic techniques with symbolic manipulation
of intentions
-- meta-level reasoning
Design and Evaluation of Rational Agents
-- architectural design of systems that incorporate one or more of the above
-- real-world and experimental applications of rational agency
-- evaluation criteria for rationality
Organizing/Programme Committee
--------------------
Michael P. Georgeff (Australian Artificial Intelligence Institute, Australia)
Martha Pollack (SRI International, Menlo Park, USA)
Stuart Russell (University of California at Berkeley, Berkeley, USA)
David Israel (SRI International, Menlo Park, USA)
Anand Rao (Australian Artificial Intelligence Institute, Australia)
Submission of Abstracts
-----------------------
Send Five (5) copies of an extended abstract (maximum 8 double-spaced pages)
by May 1, 1991 to the following address. Please include the e-mail address
of the main point of contact.
Michael P. Georgeff
Australian Artificial Intelligence Institute
1 Grattan Street
Carlton, Victoria, 3053
Australia
Tel: (+613) 663-7922
email: georgeff@aaii.oz.au
Review of Papers
----------------
The extended abstracts will be reviewed for significance, originality, and
relevance to the topic of the workshop. The workshop will be kept
small -- about 30 participants. A small number of participants will be
selected to present their work at the workshop.
Accepted abstracts are due by July 15, 1991. They will be duplicated and
distributed to all participants at the workshop. Acceptance of the abstracts
does not preclude later submissions of derivative papers to conferences and
journals.
Important Dates
---------------
Date for Submission: May 1, 1991
Notification to Authors: June 15, 1991
Receipt of Revised Abstracts: July 15, 1991
Preliminary Agenda
------------------
Session 1 (09:00 - 10:30): Foundations and Formalizations of Rational Agency
Session 2 (11:00 - 12:30): Role of Decision-Theoretic Techniques
Session 3 (14:00 - 15:30): Design and Evaluation of Rational Agents
Session 4 (16:00 - 17:00): Panel Discussion: Future Directions