IJCAI 2005 Workshop on
Reasoning, Representation, and Learning in Computer Games

http://home.earthlink.net/~dwaha/research/meetings/ijcai05-rrlcgw
Call for Participation
Description
Our goal is to encourage the study, development, integration, and
evaluation of AI techniques on tasks from complex games. These
challenging performance tasks are characterized by huge search spaces,
uncertainty, opportunities for coordination/teaming, and (frequently)
multi-agent adversarial conditions. We want to encourage dialog among
researchers in a variety of AI disciplines who seek to integrate their
approaches so as to develop and test theories on comprehensive
intelligent agents that can function competently in virtual gaming
worlds. This workshop should yield an understanding of
state-of-the-art approaches for performing well in complex gaming
environments, and research issues that require additional attention.
Topics relevant to this workshop include, but are not limited to, the
following areas:
- Knowledge intensive learning techniques for game playing
- Meta-level reasoning and representations for decision support
- Qualitative representations for simulated world states
- Reasoning in dynamic physical worlds
- Interface standards for developing AI/gaming systems
- Tools for integrating AI techniques and gaming simulators
- Explanatory AI techniques
- AI techniques in training simulators
- Adversary and adversarial strategy recognition and modeling
- Planning algorithms for large-scale environments
- Interactive tools for situation assessment
- Multi-agent game AI coordination and collaboration
- Social reasoning among virtual gaming agents
- Representations for planning and decision making
- Natural language generation/interpretation in games
- Simulating character emotions
- Dynamic story/gameplay generation and control
- Analyses of AI techniques on complex gaming tasks
- Lessons learned in AI/gaming investigations
- Surveys on state-of-the-art AI/gaming approaches
Complex games make good environments for studying AI-complete
research, and appropriate simulators can serve as excellent platforms
for posing challenging AI-hard problems. These environments can be
used to focus research and development for a broad spectrum of AI
techniques. We have particular interest in games with large search
spaces (e.g., real-time and turn-based strategy games, role-playing
games), which are challenging for current-generation AI approaches,
and usually have built-in metrics/measures for evaluation. However,
while complex gaming tasks can involve a wide array of AI-interesting
problems (e.g., plan recognition, learning, representation, logical
reasoning, search, natural language understanding), dialog among these
communities is challenging due to the fractioning of sub-disciplines.
Focused forums are needed to encourage communication among researchers
interested in this topic.
Plans
We will include an invited speaker to summarize recent work in this
area, and will also invite papers from a small but diverse set of
contributors to this area. The Organizing Committee will select a
subset of the submitted papers for oral presentation. In addition, we
will arrange a panel that addresses popular interests among the
participants. Finally, time will be reserved for a demonstration
session.
Unfortunately, conducting AI research in these environments is
difficult because integrating AI systems with gaming simulators can be
a time-intensive process. Therefore, we will define and implement a
small number of challenge problems in gaming simulators integrated
with TIELT, a freely available
test bed that researchers can use to access these simulators (along
with some integrated AI systems). Participants are welcome to address
these problems in their submissions, or to use TIELT as needed to
otherwise support their research goals. (Depending on the level of
participation, we may group contributions that involve these challenge
problems.)
Participation Process
This one-day workshop will be held on 31 July 2005 as part of the IJCAI 2005 workshop series
in Edinburgh, Scotland. This workshop is open to all (i.e., senior
and junior) members of the AI community. However, the number of
participants is limited to 40. Also, participants must register for
this workshop to attend, although they do not need to register for the
entire IJCAI'05 conference.
Our Workshop Organizing Committee will select participants through a
review process. Paper submissions should be formatted according to the
IJCAI
2005 formatting guidelines, but please include your name(s),
affiliation(s), and email address(es) at the top of the first page as
this will not be a double-blind reviewing process. Submissions should
not exceed 6 pages in length, and be in PDF format. We also welcome
short (max 4 pages) submissions concerning system demonstrations
relevant to the workshop focus. Finally, we request a 1-page
Statement of Interest from anyone who wishes to attend without
submitting a paper. In these Statements, please describe your
relevant interest, related projects (if any), and list a few
relevant publications (if any). Please email all Submissions,
Statements, or requests to be on this workshop's (moderated) mailing
list to co-chair David
W. Aha.
Important Dates
8 April 2005: Paper submissions
2 May 2005 (note revised date!): Accept/reject decisions on submitted papers
20 May 2005: Submission of camera-ready papers
31 July 2005: Workshop date! (Edinburgh, Scotland)
David W. Aha, Naval
Research Laboratory (USA) (Co-Chair)
Daniel Borrajo,
Universidad Carlos III de Madrid (Spain)
Michael Buro,
University of Alberta (Canada)
Pádraig
Cunningham, Trinity College Dublin (Ireland)
Dan Fu, Stottler-Henke
Associates, Inc. (USA)
Joahnnes
Fürnkranz, TU Darmstadt (Germany)
Joseph
Giampapa, Carnegie Mellon University (USA)
Héctor
Muņoz-Avila, Lehigh University (USA) (Co-Chair)
Alexander Nareyek,
AI Center (Germany)
Jeff Orkin, Monolith Productions
(USA)
Marc Ponsen,
Lehigh University (USA)
Pieter Spronck,
Universiteit Maastricht (Netherlands)
Michael van Lent,
University of Southern California (USA) (Co-Chair)
Ian Watson,
University of Auckland (New Zealand)
WWW: http://home.earthlink.net/~dwaha/research/meetings/ijcai05-rrlcgw