POMDP Reading Group
Spring 2009
Meetings on Monday 4 - 6pm, at Adaptive Computing Lab, COM1.
This is the list of papers that have been proposed for reading and which have not been scheduled yet. The papers that have been scheduled follow below.
Continuous POMDPs
(refer to schedule below)
Factored and other compact state-space representations
Tractable Inference for Complex Stochastic Processes
Xavier Boyen and Daphne Koller. Uncertainty in Artificial Intelligence UAI '98, 1998.
Approximate Planning for Factored POMDPs
Z. Feng and E. Hansen. Sixth European Conference on Planning (ECP-01), Toledo, Spain, September 2001.
Dynamic Programming for POMDPs using a Factored State Representation
E. Hansen and Z. Feng. Fifth International Conference on Artificial Intelligence Planning and
Scheduling (AIPS-00), Breckenridge, Colorado, April 2000.
Computing Optimal Policies for Partially Observable Decision Processes using Compact Representations
Craig Boutilier, David Poole. Proceedings of the Thirteenth National Conference on Artificial Intelligence (AAAI-96)
Value-directed Compression of POMDPs
Pascal Poupart and Craig Boutilier. Advances in Neural Information Processing Systems 15 (NIPS),
pages 1547-1554, Vancouver, BC, 2002.
Finding Approximate POMDP solutions Through Belief Compression
N. Roy, G. Gordon and S. Thrun. Journal of Artificial Intelligence Research, 23: 1-40, 2005.
(alternative to the above)
Exponential Family PCA for Belief Compression in POMDPs
N. Roy and G. Gordon. Advances in Neural Information Processing (15) NIPS, Vancouver, BC. Dec. 2002
A Novel Orthogonal NMF-Based Belief Compression for POMDPs
Xin Li, William K. W. Cheung, Jiming Liu, Zhili Wu. ICML 2007.
Hierarchical
Approximate Planning with Hierarchical Partially Observable Markov Decision Processes for Robot Navigation
Georgios Theocharous and Sridhar Mahadevan. IEEE Conference on Robotics and Automation (ICRA), Washington, D.C. May 2002.
Synthesis of hierarchical finite-state controllers for POMDPs
E. Hansen, R. Zhou. Proceedings of the Thirteenth International Conference on Automated Planning and Scheduling. 2003.
HQ-Learning
Wiering and Schmidhuber. Adaptive Behavior.1997; 6: 219-246.
Model minimization in Markov decision processes
T. Dean, R. Givan. Proceedings of the Fourteenth National Conference on Artificial Intelligence, AAAI, 1997, pp. 106-111.
Applications
(refer to schedule below)
Interactive-POMDP
A Framework for Sequential Planning in Multiagent Settings
Piotr Gmytrasiewicz and Prashant Doshi, Journal of AI Research, Vol 24: 49-79, 2005.
(or conference version:
A Framework for Sequential Planning in Multi-Agent Settings
Piotr Gmytrasiewicz and Prashant Doshi, Proceedings of the Eighth International Symposium on Artificial Intelligence and Mathematics, January 4-6, 2004.)
Decentralized-POMDP
(refer to schedule below)
Meeting Schedule
| Date
| Topic
| Presenter
| Remarks
|
28 Feb
08
| Applications
Assisting Persons with Dementia during Handwashing Using a Partially Observable Markov Decision Process
Jesse Hoey, Axel von Bertoldi, Pascal Poupart, and Alex Mihailidis. Proceedings of the Intl Conf on Vision Systems (ICVS), Biefeld, Germany, 2007.
Value-Directed Human Behavior Analysis from Video Using Partially Observable Markov Decision Processes
Jesse Hoey and James J. Little. IEEE PAMI, Vol 29, No 7, pp.1118-1132, July 2007.
| Sylvie Ong
| Discuss papers to be scheduled for upcoming meetings
|
13 Mar
08
| Factored Belief Representations
Solving Factored POMDPs with Linear Value Functions
Carlos Guestrin, Daphne Koller, Ronald Parr. IJCAI-01 workshop on Planning under Uncertainty and
Incomplete Information, 2001.
Approximate Planning for Factored POMDPs using Belief State Simplification
David A. McAllester, Satinder Singh. UAI 99.
| Lee Wee Sun
|
|
3 April
08
| Hierarchical
Policy-contingent abstraction for robust robot control
J. Pineau, G. Gordon & S. Thrun. Conference on Uncertainty in Articifical Intelligence (UAI). Acapulco, Mexico. pp. 477-484. Aug. 2003.
Automated Hierarchy Discovery for Planning in Partially Observable Environments
Laurent Charlin, Pascal Poupart and Romy Shioda. Advances in Neural Information Processing Systems 19 (NIPS), Vancouver, BC, 2006.
| Amit
| Discuss papers to be scheduled for upcoming meetings
|
24 April
08
| Decentralized POMDPs
Bounded Policy Iteration for Decentralized POMDPs
Daniel S. Bernstein, Eric A. Hansen, and Shlomo Zilberstein. Proceedings of the 19th International Joint Conference on Artificial Intelligence (IJCAI-05), Edinburgh, Scotland, July 2005.
Improved Memory-Bounded Dynamic Programming for Decentralized POMDPs
Sven Seuken and Shlomo Zilberstein. Proceedings of the Twenty-Third Conference on Uncertainty in
Artificial Intelligences (UAI-07), Vancouver, BC, Canada, July, 2007.
| Hanna Kur
| Discuss papers to be scheduled for upcoming meetings
|
08 May
08
| Continuous POMDPs
Point-based Value Iteration for Continuous POMDPs
Josep M. Porta, Nikos Vlassis, Matthijs T.J. Spaan, Pascal Poupart. Journal of Machine Learning
Research 7(Nov):2329--2367, 2006.
Parametric POMDPs for Planning in Continuous State Spaces
A. Brooks, A. Makarenko, S.B. Williams, & H.F. Durrant-Whyte. Robotics
and Autonomous Systems, vol. 54, no. 11, 2006, pp. 887-897.
| David Hsu
| Discuss papers to be scheduled for upcoming meetings
|
29 May
08
| Policy Iteration
Solving POMDPs by Searching in Policy Space
Eric Hansen. UAI-98.
Point-Based Policy Iteration.
Shihao Ji, Ronald Parr et al. AAAI 2007.
| Hanna Kur
|
|
19 June
08
| Policy Iteration
Solving POMDPs by Searching the Space of Finite Policies
Nicolas Meuleau, Kee-Eung Kim, Leslie Pack Kaelbling, Anthony R. Cassandra. Proc. of the Conf. on Uncertainty in AI, 1999.
PEGASUS: A policy search method for large MDPs and POMDPs.
Andrew Y. Ng, Michael Jordan. Proc. of the Conf. on Uncertainty in AI, 2000.
| Sylvie Ong
|
|
16 Feb
09
| ADDs part 1
Symbolic Boolean Manipulation with Ordered Binary Decision Diagrams
R. E. Bryant. ACM Computing Surveys, Vol. 24, No. 3 (September, 1992), pp. 293–318.
Algebraic Decision Diagrams and Their Applications
R.I. Bahar; E.A. Frohm; C.M. Gaona; G.D. Hachtel; E. Macii; and A. Pardo.
International Conference on CAD (1993), 188–191.
Graph-based algorithms for boolean function manipulation.
R. E. Bryant. IEEE Transactions on Computers 35(8):677–691, 1986.
Efficient ADD Operations for Point-Based Algorithms
Guy Shani, Ronen I. Brafman, Solomon E. Shimony and Pascal Poupart.
Proceedings of the Eighteenth International Conference on Automated Planning and Scheduling (ICAPS),
Sydney, Australia, 2008.
|
|
|
Adaptive Intelligence Reading Group
Spring 2005
The semester's reading is centered around techniques to allow a mobile robot to automatically map the environment, recognize human activities and decide the appropriate action to achieve its objectives.
| Date
| Topic
| Presenter
| Remarks
|
| 18 March
| Estimate where you are: Kalman filtering
| Amit
|
|
| 1 April
| Estimate where you are: Particle filtering
| Nikhila
|
|
| 15 April
| Estimate where you are: Expectation Maximization (EM)
| Wee Sun
| The method is essentially running EM on a HMM. Those not familiar with HMM may want to have a look at the HMM notes from 29 April.
|
| 29 April
| Recognizing activities: Hidden Markov Model (HMM), Conditional Random Field (CRF )
| Hai Leong
|
|
| 13 May
| Decide what you do: MDP and POMDP
| Hanna
|
|
| 27 May
| Decide what you do: on-line decision making
| Xiaoxi
|
|
