Modeling Other Agents from Observations (MOO 2004)
A Workshop at the International Joint Conference on Autonomous Agents and Multi-Agent Systems
Mon, July 19, 2004, Columbia University, NY, USA
A Workshop at the International Joint Conference on Autonomous Agents and Multi-Agent Systems
Mon, July 19, 2004, Columbia University, NY, USA
Modeling other agents from observations involves learning and reasoning about agents based only on observations of their behavior, i.e., their interaction with the environment and with each other. The observed agents may be synthetic software agents, physical robots, or humans.
Given observations of the actions (or other observable features) of agents, a modeling agent's task is to infer their beliefs, plans, intentions, goals, etc. This is a synergistic area of research in autonomous agents and multi-agent systems, combining and unifying techniques of plan recognition, user modeling, intelligent user interfaces, human/computer interaction, natural language understanding, machine learning, and intention recognition. Agent modeling plays a crucial role in application areas ranging from e-commerce and collaborative filtering, to software assistants, to observation-based coordination in robots and software agents, to overhearing and imitation. This wide-spread diversity of applications and disciplines, while producing a wealth of ideas and results, has unfortunately contributed to fragmentation in the field, as researchers publish relevant results in a wide spectrum of journals and conferences.
This workshop seeks to bring together researchers and practitioners of agent modeling from diverse backgrounds, to share in ideas and recent results. It will aim to identify important research directions and to identify opportunities for synthesis and unification. The workshop will especially emphasize a discussion of challenges and trends: Reasoning with incomplete or incorrect knowledge of others, agent modeling on the web and in open multi-agent systems, modeling multiple agents and the interactions between them, inferring the capabilities of agents from observations, and using agent modeling for coordination and teamwork. Contributions are sought in the following areas of research:
Due to the diversity of disciplines engaging in modeling agents based
on observations, relevant contributions in other fields are also welcome.
The one-day workshop will consist of a series of research presentations, organized into topical sessions (topics to be decided based on submissions). An interdisciplinary panel is planned, seeking to highlight research contributions and challenges unifying and differentiating the different sub-areas.
We are currently examining post-workshop publication venues, such as a journal special-issue and/or edited volume.
We seek novel research contributions in all areas of modeling other agents. Syntheses and survey papers are welcome, as long as they offer a fresh perspective on the field.
Submissions should follow the AAMAS-04 paper format. Authors should submit papers electronically (in PS or PDF format) to Gal Kaminka (galk@cs.biu.ac.il). The subject-line of the submission should be "MOO-04 Submission".