AGENTS-2001 Workshop
on
LEARNING AGENTS
May 29, 2001
Montreal, Canada
Organizing committee members:
- Doina Precup
(Co-Chair), McGill University, dprecup@cs.mcgill.ca
- Peter Stone
(Co-Chair), AT&T Labs -- Research, pstone@cs.cmu.edu
- Hans-Dieter Burkhard, Humboldt University, hdb@informatik.hu-berlin.de
- Amy Greenwald, Brown Univeristy, amygreen@cs.brown.edu
- Tim Oates, MIT AI Lab, oates@ai.mit.edu
- Enric Plaza, IIIA-Spanish Scientific Research Council, enric@iiia.csic.es
- Sandip Sen, University of Tulsa, sandip@kolkata.mcs.utulsa.edu
- Kagan Tumer, NASA Ames Research Center, kagan@ptolemy.arc.nasa.gov
- Moshe Tennenholtz, Technion/Stanford University, moshe@robotics.stanford.edu
- Eiji Uchibe, Osaka University, uchibe@er.ams.eng.osaka-u.ac.jp
- Manuela Veloso, Carnegie Mellon University, veloso@cs.cmu.edu
- Jose Vidal, University of South Carolina, vidal@sc.edu
DESCRIPTION
Intelligent agents often work in environments which are at best
partially understood and where the domain characteristics or
participants change over time. Under such circumstances, learning and
adaptation are key for obtaining good performance. In addition,
agents can serve their associate users much more effectively if they
are able to capture the unspecified and/or changing preferences of
these users.
Another aspect of agent based systems is that they often are situated
in a multiagent environment. Agents in such systems have to interact
both with associated users and other agents. Coordination of the
activities of multiple agents, whether selfish or cooperative, is
essential for the viability of any system in which multiple agents
must coexist. Learning and adaptation are invaluable mechanisms by
which agents can evolve coordination strategies that meet the demands
of the environment and the requirements of individual agents.
The goal of this workshop is to focus on research addressing the
unique requirements that autonomous agents impose on learning methods.
The Learning Agents workshop organized jointly at the Agents'2000 and
ECML'2000 conferences started a fruitful discussion between
researchers involved in designing and applying machine learning
techniques to autonomous agents. The workshop attracted more than 40
researchers studying learning agents. The workshop ended with a
lively discussion regarding the special properties of agent learning
as opposed to machine learning in general. The proposed workshop aims
to continue and extend the discussion started last year.
Topics of interest
We especially encouraged the submission of papers addressing the
following topics:
- Benefits of adaptive/learning agents over agents with fixed
behavior.
- Evaluation of the effectiveness of individual learning strategies
(e.g., case-based, explanation-based, inductive, reinforcement
learning), or multistrategy combinations.
- Characterization of learning and adaptation methods in terms of
modeling power, communication abilities, knowledge requirements and
processing abilities of individual agents.
- Developing learning and adaptation strategies, or reward
structures, for environments with cooperative agents, selfish agents,
partially cooperative agents (agents that will cooperate only if
individual goals are not sacrificed) and for environments that can
contain mixture of these types of agents.
- Analyzing and constructing algorithms that guarantee the
convergence and stability of group behavior in multi-agent systems.
- Analyzing the effects of knowledge acquisition mechanisms on the
responsiveness of agents or groups to the addition/deletion of other
agents from the environment.
- Agents learning by observing users or other agents.
- Evolving agent behaviors or co-evolving multiple agents with
similar/opposing interests.
- Investigation of teacher-student relationships between agents and
users.
Submission Requirements (past due)
E-mail the URL of either a
- brief statement of interest (1 page), or
- complete paper (3000 words maximum) including keywords and
authors' complete address to Peter Stone and Doina Precup at
pstone@cs.cmu.edu and dprecup@cs.mcgill.ca.
Papers and statements of interest must be in one of the following
formats: postscript, pdf, HTML.
Direct all questions and inquiries to:
Doina Precup (Co-chair)
School of Computer Science
McGill University
3480 University st.
Montreal, Quebec, Canada
H2A 1A7
(514) 398-6443
(514) 398-3883 (fax)
dprecup@cs.mcgill.ca
http://www.cs.mcgill.ca/~dprecup
Important Dates
Workshop: May 29, 2001