AAAI 2004 Fall Symposium on
Artificial Multiagent Learning

To be Held October 21-24, 2004, Washington D.C.
Part of The AAAI 2004 Fall Symposium Series
Schedule In Pretty PDF Format. We are strongly considering setting aside a session (likely the last one on Sunday) for any attendies to do quick spiels about their multiagent systems and resarch. If you're interested, please send email to Sean Luke (sean -- at --

Barring no other proposed activity, we are suggesting that the special activity be a trip to either the Smithsonian Museum of the American Indian or (a long jaunt, but probably worth it) the Smithsonian National Air and Museum Udvar Hazy Center. We'll decide at the workshop.

Friday Oct 22
  • Adapting Network Structure for Efficient Team Formation
    Matthew Gaston, John Simmons, and Marie desJardins
  • Analyzing the Effects of Tags on Promoting Cooperation in Prisoner's Dilemma
    Austin McDonald and Sandip Sen
  • Co-Evolving Team Capture Strategies for Dissimilar Robots
    H. Joseph Blumenthal and Gary Parker
  • Stochastic Direct Reinforcement: Application to Simple Games with Recurrence
    John Moody, Yufeng Liu, Matthew Saffell, and Kyoungju Youn
  • Dynamics of Strategy Distribution in Iterated Games
    Stephane Airiau, Sabyasachi Saha, and Sandip Sen
Saturday Oct 23
  • Empirical Comparison of Incremental Learning Strategies for Genetic Programming-Based Keep-Away Soccer Agents
    Scott Harmon, Edwin Rodriguez, Christopher Zhong, and William Hsu
  • Evolving Control for Micro Aerial Vehicles (MAVs)
    M. Rhodes, G. Tener, and Annie Wu
  • Learning e-Pareto Efficient Solutions with Minimal Knowledge Requirements Using Satisficing
    Jacob Crandall and Michael Goodrich
  • Opportunities for Learning in Mulit-Agent Meeting Scheduling
    Elisabeth Crawford and Manuela Veloso
  • Safe Strategies for Agent Modelling in Games
    Peter McCracken and Michael Bowling
  • Understanding Competitive Co-evolutionary Dynamics via Fitness Landscapes
    Elena Popovici and Kenneth De Jong
  • Learning TOMs: Towards Non-Myopic Equilibria
    Arjita Ghosh and Sandip Sen
  • Multi-Agent Learning in Conflicting Multi-level Games with Incomplete Information
    Maarten Peeters, Katja Verbeeck, and Ann Nowe
  • Multi-agent Learning in Mobilized Ad-Hoc Networks
    Yu-Han Chang and Leslie Pack Kaelbling
Sunday Oct 23
  • On the Agenda(s) of Research on Multi-Agent Learning
    Robert Powers and Yoav Shoham
  • Tags and the Evolution of Cooperation in Complex Environments
    Lee Spector, Jon Klein, and Chris Perry
  • Learning Payoff Functions in Infinite Games
    Yevgeniy Vorobeychik, Michael Wellman, and Satinder Singh
Final Copy Instructions Final print copy of papers is due on August 31. By that date, please submit:
  • Signed "Permission to Distribute" form
  • Audio/Video request form (if any)
  • A camera-ready copy of the paper to AAAI

Follow the following instructions to download forms and submit your paper. The paper must be formatted according to AAAI guidelines. Individual forms and instructions can also be found on the bottom of this page

Topic Multiagent systems is a subset of distributed artificial intelligence that emphasizes the joint behaviors of agents in environments with some degree of autonomy. In most such environments there are constraints placed on the degree to which any agent may know what other agents know, or on their communication capabilities, such that the system must have distributed control and cannot be solved with a master-slave model via a single master agent.

In recent years there has been increasing interest in applying machine learning techniques to multiagent systems problems. The presence of large numbers of agents, increasingly complex agent behaviors, partially observable environments, and the mutual adaptation of agent behaviors make the learning process a challenging one. These problems are further complicated by noisy sensor data, local bandwidth-limited communication, unplanned faults in hardware agents, and stochastic environments.

The goal of this symposium is to bring together researchers from diverse areas of the multiagent learning community. Because the dynamics of multiagent learning are such that the correct answer is often not known beforehand, much of the multiagent learning research to date has focused on reinforcement learning or stochastic optimization (evolutionary computation, ant colony optimization, etc.). Some topics of interest include, but are not restricted to:

  • Coevolution and coadaptation
  • Adversarial methods for learning game-playing strategies
  • Swarm and social learning methods
  • Evolutionary game theory
  • Multirobot learning
  • Effects of communication on learning
  • Discovery of emergent behavior
  • Automatic formation of coalitions, contracts, and markets
  • Game-theoretic issues in multiagent learning
  • Agent modeling
Organizing Committee
  • Michael Bowling (bowling @
  • Kenneth De Jong (kdejong @
  • Marie desJardins (mariedj @
  • Sean Luke (chair, sean @
  • Mitchell Potter (mpotter @
  • Lee Spector (lspector @
Related Symposia We are considering a joint session of some sort with the Real-Life Reinforcement Learning Symposium. For other symposia, see the AAAI Symposia Website.