Title: Towards Conquering Uncertainty in Agent Systems
Speaker: Janusz Marecki, Mathematical Sciences Dept, IBM T. J. Watson Research.
Date: Monday, May 24, 2010 11:00 am- 12:00 pm****
Location: DIMACS Center, CoRE Bldg, Room 431, Rutgers University, Busch Campus, Piscataway, NJ
****Please note time change
In most domains involving either individual agents or teams, ranging from autonomous space exploration to automated personal assistants to sensor networks, intelligent agents must fundamentally reason with uncertainty. In providing techniques for reasoning with such uncertainty, we must address at least three major challenges: (i) these techniques handle both discrete and continuous features of the environment (ii) they are applicable to partially observable environments and (ii) they scale up in the number of agents, yet provide quality guarantees, as ad-hoc approaches without guarantees may cause significant failures in serious applications. In this talk I will present two key ideas that help to address these challenges: (i) An analytic algorithm for solving Markov Decision Processes (MDPs) with continuous resources that is orders of magnitude faster than its competitors and (ii) An algorithm for solving Distributed, Partially Observable MDPs that exploits agent interaction structure and varies agent policy expressivity to scale up the number of agents to double digits. Furthermore, I will demonstrate the practical benefit of my algorithms on the deployed systems and outline my vision of the Agent Systems of tomorrow.
Janusz Marecki is a research staff member at the mathematical sciences department at IBM T.J. Watson Research Center. Janusz obtained his Ph.D in artificial intelligence form from University of Southern California and Dr.Sc in mathematical modeling from State Scientific and Research Institute of Information Infrastructure in Ukraine. Prior to joining IBM Research, Janusz was a research assistant at the European Laboratory for Nuclear Research, a research associate at the Ukrainian Academy of Sciences and a lecturer at the Academy of Computer Sciences in Poland. His research interests are in reasoning under uncertainty in single/multiagent systems with an emphasis on planning with continuous resources in stochastic environments. He is an author of over 60 refereed publications and four patents.