POMDP Symposium Presentation Page

• Drew Bagnell, Nechyba, A. Antonio Arroyo (UFL). Sessions: Robotic Applications. Paper: Comparison of Reinforcement Learning Techniques for Automatic Behavior Programming by J. A. Bagnell, K. L. Doty, A. A. Arroyo (6 pages). Status: Bagnell confirmed, registered; Arroyo registered.
• Valentina Bayer (Oregon St.). Sessions: Approximations and Heuristics. Paper: Cost-observable MDPs by V. Bayer (~1 page). Status: confirmed, registered.
• Craig Boutilier (UBC). Sessions: Factored Models (overview), Factored Models. Paper: Computing Optimal Policies for Partially Observable Decision Processes using Compact Representations by C. Boutilier, D. Poole (8 pages). Status: confirmed, not registered. Constraints: also presenting at another symposium (time unknown).
• Tony Cassandra (MCC). Sessions: Other Applications (overview). Paper: A Survey of POMDP Applications by A. R. Cassandra (8 pages). Status: confirmed, registered.
• Jay Chawla (Maryland). Sessions: Mathematical and Algorithmic Foundations. Paper: Analysis of a Risk Sensitive Control Problem for Hidden Markov Chains (13 pages, addendum), Risk-sensitive Optimal Control of Hidden Markov Models: Structural Results by E. Fernandez-Gaucherand, S. I. Marcus (13 pages). Status: confirmed, registered.
• Trevor Darrell (Interval). Sessions: Other Applications. Paper: Reinforcement Learning of Active Recognition Behaviors by T. Darrell (17 pages). Status: confirmed, registered.
• Sascha E. Engelbrecht (Umass). Sessions: Extensions and Specializations. Paper: Planning with Delayed State information by S. E. Engelbrecht, K. V. Katsikopoulos (4 pages). Status: confirmed, registered.
• Hector Geffner, Blai Bonet (Simon Bolivar). Sessions: Other Applications, Approximations and Heuristics. Paper: Solving Large POMDPs using Real Time Dynamic Programming by H. Geffner, B. Bonet (8 pages). Status: confirmed, registered. Constraints: Hector busy day 2 from 11am-3pm.
• Judy Goldsmith, Tong Li, Chris Wells, Shelia Sittinger (U. Kentucky). Sessions: Mathematical and Algorithmic Foundations. Paper: Complexity Issues in Markov Decision Processes by J. Goldsmith, M. Mundhenk (9 pages). Status: confirmed, registered.
• Eric Hansen (Mississippi St.). Sessions: Mathematical and Algorithmic Foundations. Paper: Solving POMDPs by Searching in Policy Space by E. Hansen (9 pages). Status: confirmed, registered.
• Milos Hauskrecht. Sessions: Approximations and Heuristics (overview), Other Applications. Paper: Planning Medical Therapy Using Partially Observable Markov Decision Processes (8 pages, in addendum), Incremental Methods for Computing Bounds in Partially Observable Markov Decision Processes (6 pages, in addendum), Statement of Interest by M. Hauskrecht (1 page). Status: confirmed, registered.
• Leslie Pack Kaelbling (Brown). Sessions: Factored Models. Paper: Learning Topological Maps with Weak Local Odometric Information (8 pages, in addendum). Status: confirmed, registered.
• Steve Kemp, Dave Eckerman (UNC). Sessions: Other Applications. Paper: Language of Animal Learning Theories: A radical behaviorist perspective (14? pages, in addendum). Status: confirmed, registered.
• Pier Luca Lanzi (Politecnico di Milano). Sessions: Reinforcement Learning. Paper: Adding Memory to Wilson's XCS Classifier System to Learn in Partially Observable Environments by P. L. Lanzi (8 pages). Status: confirmed, registered.
• Michael Littman (Duke). Sessions: Introduction, Extensions and Specializations (overview), Approximations and Heuristics. Paper: Chapter 6: Partially Observable markov Decision Processes (excerpt from Algorithms for Sequential Decision Making) (25 pages, in addendum). Status: confirmed, registered.
• Chris Lusena (U. Kentucky). Approximations and Heuristics. Paper: Nonapproximability Results for Markov Decision Processes by C. Lusena, J. Goldsmith, M. Mundhenk (12 pages). Status: confirmed, registered.
• Omid Madani (Washington). Sessions: Mathematical and Algorithmic Foundations. Paper: Computational Models for Decision Making in Dynamic and Uncertain Domains (27 pages, in addendum), On the Computability of Infinite Horizon Partially Observable Markov Decision Processes (24 pages, in addendum), also Abstract by O. Madani (~1 page). Status: confirmed, registered.
• Sridhar Mahadevan (MSU). Sessions: Robotic Applications (overview), Robotic Applications, Other Applications. Paper: Partially Observable Semi-Markov Decision Processes: Theory and Applications in Engineering and Cognitive Science by S. Mahdevan (8 pages). Status: confirmed, registered. Constraints: coming late on day 1.
• Steve Majercik (Duke). Sessions: Factored Models. Paper: MAXPLAN: A New Approach to Probabilistic Planning by S. M. Majercik, M. L. Littman (8 pages). Status: confirmed, registered.
• Niels Peek (Utrecht). Sessions: Other Applications. Paper: Predictive Probabilistic Models for Treatment Planning in Paediatric Cardiology by N. Peek (6 pages). Status: confirmed, registered.
• Loren Platzman (self). Sessions: Introduction, Approximations and Heuristics. Paper: A Feasible Computational Approach to Infinite-horizon Partially-observed Markov Decision Problems (53 pages, in addendum, not online), POMDP Research in the Seventies (3 pages, in addendum). Status: confirmed, registered.
• Larry Pyeatt (Colorado St.). Sessions: Reinforcement Learning. Paper: (17 pages, in addendum). Status: confirmed, registered.
• Nicholas Roy (CMU). Sessions: Robotic Applications. Paper: Coastal Navigation---Robot Motion with Uncertainty by N. Roy, W. Burgard, D. Fox, S. Thrun(6 pages). Status: confirmed, registered.
• Satinder Singh (AT&T). Sessions: Reinforcement Learning (overview?), Reinforcement Learning. Paper: Using Eligibility Traces to Find the Best Memoryless Policy in Partially Observable Markov Decision Processes by J. Loch, S. Singh (9 pages). Status: confirmed, registered.
• Claude Touzet (CESAR-ORNL). Sessions: Robotic Applications. Paper: Cooperative Reinforcement Learning and Hidden Markov Models by C. Touzet (1 page). Status: confirmed, registered.
• Rich Washington (NASA). Sessions: Approximations and Heuristics. Paper: BI-POMDP: Bounded, Incremental Partially-Observable Markov-Model Planning by R. Washington (12 pages). Status: confirmed, registered.
• Chip White (Michigan). Sessions: Approximations and Heuristics. Paper: Nonhomogeneous Markov Decision Processes by J. C. Bean, C. C. White, III, Z-Z. Lin (~3 pages). Status: confirmed, not registered. Constraints: only around for day 2.
• Nevin Zhang, Weihong Zhang (HK UST). Sessions: Mathematical and Algorithmic Foundations (overview), Mathematical and Algorithmic Foundations. Paper: Planning with Partially Observable Markov Decision Processes: Advanced in Exact Solution Method by N. L. Zhang, S. S. Lee (8 pages). Status: confirmed, registered. Constraints: Nevin probably not around on day 3.

Michael L. Littman, Ph.D.

Assistant Professor

Department of Computer Science

Duke University, Durham, NC 27708-0129

Office: D209 LSRC

Phone: 919-660-6537

Fax: 919-660-6519

mlittman@cs.duke.edu