(please see the main page for schedule information)
For the past decade, the TEAMCORE research group has focused on building agent teams for complex, dynamic domains. Early work, inspired by belief-desire-intention (BDI) logics and explicit team plans, has led to public-domain, reusable team- coordination infrastructures. More recently, however, our focus shifted to distributed constraint optimization and distributed POMDPs in building agent teams. In this context, I will present Multiagent Team Decision Problems (MTDPs), that are based on distributed, communicating POMDPs. MTDPs provide us complexity results for key types of teamwork domains, and allow us to compare complexity and optimality of competing team coordination algorithms. In addition, we have designed algorithms that enable efficient computation of (locally) optimal policies for MTDPs. Despite the promise of the MTDPs, scalability remains problematic. We propose a novel solution based on creating a hybrid BDI-MTDP approach, that has provided promising results. In particular, BDI team plans are exploited to improve MTDP tractability and MTDPs improve BDI team plan performance. I will present some results from applying this approach in the RoboCup-Rescue domain. Research reported conducted in collaboration with the TEAMCORE research group (http://teamcore.usc.edu).
Milind Tambe is an Associate Professor of Computer Science at University of Southern California(USC). He received his Ph.D. from the School of Computer Science at Carnegie Mellon University. He leads the TEAMCORE research group at USC, with research interests in multi-agent systems, specifically multi-agent teamwork, adjustable autonomy and distributed negotiations. He was general co-chair for the International Joint Conference on Agents and Multiagent Systems (AAMAS) 2004, and was chair of the organizing committee for the first Americas Agents School and program co-chair of the International conf on multi- agent systems (ICMAS) 2000. He is also currently associate editor of the Journal of Artificial Intelligence Research (JAIR), and on the editorial board of the Journal of Autonomous Agents and Multi-agent Systems (AAMAS). A current member of the board of directors of the International foundation for multiagent systems, he has also served on the board of trustees of RoboCup, the Robot World Cup Federation. Some recent awards include the Okawa foundation research grant (2003), AAMAS best paper award (2002) and the RoboCup scientific challenge award (1999).