Re: Proposed agent modelling workshop

Nick Jennings (N.R.Jennings@qmw.ac.uk)
Fri, 21 Oct 1994 16:25:04 +0100 (BST)

In light of this mornings email message about an IJCAI-95 workshop
proposal by Lewis Johnson, we thought it would be appropriate also to
state publically that we are intending to submit a proposal on a very
much related topic - namely Agent Theories, Architectures & Languages.
This proposal will be similar to the very successful workshop which we
organised at this years ECAI conference and the draft of this proposal is
given below.

AGENT THEORIES, ARCHITECTURES, AND LANGUAGES
An IJCAI-95 Workshop
[Draft Only - Draft Only - Draft Only]

Introduction

Artificial Intelligence is ultimately about building intelligent agents,
and yet it is only comparatively recently - since about the mid 1980s -
that issues surrounding the synthesis of intelligent autonomous agents have
entered the mainstream of AI. Since then, however, there has been an
intense flowering of interest in the subject: agent-based computing is now
one of the most rapidly expanding areas of both AI and computer science
generally.

The aim of this workshop is to address all issues related to the theory and
practice of intelligent agents. Thus the workshop will address such issues
as the specification of agents via agent theories, agent architectures and
methodologies for realising agents, agent languages and software tools for
programming and experimenting with agents, and novel applications of agent
technology. Issues such as agent communication languages also fall within
the scope of the workshop. However, the submission of papers that address
mainstream Distributed AI issues (such as cooperative problem solving or
cooperation protocols) is not encouraged, as there are other more appropri-
ate outlets for such work.

The 1995 workshop will build on the success of a workshop with the same
title that was held at the 1994 European conference on AI. The proceedings
of this workshop are to be published in early 1995 by Springer-Verlag, in
their `Lecture Notes in AI' series, under the title `Intelligent Agents'.
It is hoped that the proceedings of the 1995 workshop will be published in
a similar way.

As the title suggests, the workshop has three main themes:

o Agent theories: How do the various components of an agent's cognitive
makeup conspire to produce rational behaviour? What is the relation-
ship between these components? What formalisms are appropriate for
expressing aspects of agent theory? Do we need logic-based formalisms?
If not, is another type of mathematical framework appropriate? How are
we to model bounded rationality? What properties are desirable for an
agent communication language?

o Agent architectures: What structure should an agent have? Is reactive
behaviour enough, or do we need deliberation as well? How can we
integrate reactive and deliberative components cleanly? What is the
relationship between an agent theory and architecture? How can we syn-
thesise an agent from an agent specification? How are we to reason
about reactive systems?

o Agent languages: What are the right primitives for programming an
intelligent agent? How are these primitives related to the theory of
an agent, or its architecture? Can we realistically hope to execute
agent specifications in complex, perhaps multi-modal languages?

Papers that cross theme boundaries are of particular interest. Examples
might include a paper that demonstrated how a particular architecture or
language embodied some theory of agency, or a paper that gave the semantics
for an implemented agent communication language.

Topics of Interest

Topics of interest include, but are by no means restricted to the follow-
ing:

Agent Theories Agent Architectures

intentions deliberative architectures
time, desires, beliefs, and goals reactive architectures
situated automata theory hybrid architectures
believable agents BDI architectures
specification/verification of agents
executing logical agent specifications Agent Languages
rationality & bounded rationality
agent communication languages agent specification languages
agentification the agent-oriented paradigm
agent-based computing
Novel applications of agent technology

Organising Committee

Michael Wooldridge Nicholas Jennings
Department of Computing Department of Electronic Engineering
Manchester Metropolitan University Queen Mary & Westfield College
Chester Street Mile End Road
Manchester M1 5GD, U.K. London E1 4NS, U.K.
Email M.Wooldridge@doc.mmu.ac.uk Email N.R.Jennings@qmw.ac.uk
Tel (+44 61) 247 1531 Tel (+44 71) 975 5349
Fax (+44 61) 247 1483 Fax (+44 81) 981 0259

Program Committee

(The PC for the 1994 workshop is listed; the final PC for the 1995 workshop
is still being prepared.)

Phil Cohen (USA) Michael Fisher (UK) Piotr Gmytrasiewicz (USA)
Hans Haugeneder (Germany) Sarit Kraus (Israel) Anand Rao (Australia)
Yoav Shoham (USA) Munindar Singh (USA)

Nick Jennings &
Mike Wooldridge

/=======================================================================\
| Dr Nick Jennings |
| Department of Electronic Engineering, |
| Queen Mary & Westfield College, Phone: +44-171-975-5349 |
| University of London, Fax : +44-181-981-0259 |
| Mile End Road, Email: N.R.Jennings@qmw.ac.uk |
| London E1 4NS, UK |
| WWW page: http://www.elec.qmw.ac.uk/dai/ |
\=======================================================================/