@MASTERSTHESIS{Charlin06, author = "L. Charlin", title = "Automated Hierarchy Discovery for Planning in Partially Observable Domains", school = "University of Waterloo", year = "2006" } @techreport{MINLPBB, author = "S. Leyffer", title ="User manual for MINLP\_BB", institution = "University of Dundee, Numerical. Analysis Report", number = "NA/XXX", year = "1999" } @article{Balas:ADM79, author = {E. Balas}, title = {Disjunctive programming}, journal = {Annals of Discrete Mathematics}, volume = {5}, pages = {3--51}, year = {1979}} @article{gill-murray-saunders:siam05, author = {P. Gill and W. Murray and M. Saunders}, title = "{SNOPT}: An {SQP} Algorithm for Large-Scale Constrained Optimization", journal = {SIAM Review}, volume = {47}, number = {1}, year = {2005}, pages = {99--131}, } @inproceedings{ braziunas-boutilier:aaai04, author = "D. Braziunas and C. Boutilier", title = "Stochastic Local Search for {POMDP} Controllers", booktitle = "AAAI", pages = "690--696", year = "2004", url = "citeseer.ist.psu.edu/braziunas04stochastic.html" } @inproceedings{meuleau-peshkin-kim-kaelbling:uai99, author = "N. Meuleau and L. Peshkin and K.-E. Kim and L. Kaelbling", title = "Learning Finite-State Controllers for Partially Observable Environments", pages = "427--436", year = "1999", booktitle = "UAI", url = "citeseer.ist.psu.edu/article/meuleau99learning.html" } @inproceedings{meuleau-kim-kaelbling-cassandra:uai99, author = "N. Meuleau and K.-E. Kim and L. Kaelbling and A. Cassandra", title = "Solving {POMDPs} by Searching the Space of Finite Policies", pages = "417--426", year = "1999", booktitle = "UAI"} @inproceedings{amato07, title ="Solving {POMDPs} Using Quadratically Constrained Linear Programs", author ="C. Amato and D. Bernstein and S. Zilberstein", booktitle = "To appear In International Joint Conferences on Artificial Intelligence (IJCAI)", year = "2007" } @inproceedings{amato-bernstein-zilberstein:aima06, title= "Solving {POMDPs} Using Quadratically Constrained Linear Programs", author = "C. Amato and D. Bernstein and S. Zilberstein", booktitle = "AI and Math", year = "2006" } @Phdthesis{Poupart05, Author={Pascal Poupart}, Title="Exploiting Structure to efficiently solve large scale partially observable {Markov} decision processes", School={University of Toronto}, Year={2005} } %@inproceedings{citeulike:143527, abstract = {We develop a hierarchical approach to planning for partially observable Markov decision processes (POMDPs) in which a policy is represented as a hierarchical finite-state controller.}, author = {E. Hansen and R. Zhou}, citeulike-article-id = {143527}, comment = {Like MAXQ for FSCs and POMDPs. Learns recursively optimal hi erarchies of finite state controllers given full knowledge of the environment an d a predefined task hierarchy.}, booktitle = {ICAPS}, title = {Synthesis of Hierarchical Finite-State Controllers for POMDPs}, url = {http://citeseer.ist.psu.edu/600233.html}, year = {2003} } @inproceedings{ hansen98improved, author = "E. Hansen", title = "An Improved Policy Iteration Algorithm for Partially Observable {MDP}s", booktitle = "NIPS", year = "1998", url = "citeseer.ist.psu.edu/hansen97improved.html" } @article{ Sondik78, author = "E. Sondik", title = "The optimal control of partially observable decision processes over the infinite horizon: Discounted cost", journal = "Operations Research", volume ="26", number = "2", pages = "282--304", year = "1978" } @inproceedings{ poupart03bounded, author = "P. Poupart and C. Boutilier", title = "Bounded finite state controllers", booktitle = "NIPS", year = "2003", url = "citeseer.ist.psu.edu/poupart03bounded.html" } @inproceedings{DBLP:conf/aips/HansenZ03, author = {E. Hansen and R. Zhou}, title = "Synthesis of Hierarchical Finite-State Controllers for {POMDPs}", booktitle = {ICAPS}, year = {2003}, pages = {113-122}, crossref = {DBLP:conf/aips/2003}, bibsource = {DBLP, http://dblp.uni-trier.de} } @proceedings{DBLP:conf/aips/2003, editor = {E. Giunchiglia and N. Muscettola and D. Nau}, title = {Proceedings of the Thirteenth International Conference on Automated Planning and Scheduling (ICAPS 2003), June 9-13, 2003, Trento, Italy}, booktitle = {ICAPS}, publisher = {AAAI}, year = {2003}, isbn = {1-57735-187-8}, bibsource = {DBLP, http://dblp.uni-trier.de} } @article{280145, author = {L. Kaelbling and M. Littman and A. Cassandra}, title = {Planning and acting in partially observable stochastic domains}, journal = {Artificial Intelligence}, volume = {101}, number = {1-2}, year = {1998}, pages = {99--134} } @inproceedings{DBLP:conf/icml/GhavamzadehM03, author = {M. Ghavamzadeh and S. Mahadevan}, title = {Hierarchical Policy Gradient Algorithms.}, booktitle = {ICML}, year = {2003}, pages = {226-233}, crossref = {DBLP:conf/icml/2003}, bibsource = {DBLP, http://dblp.uni-trier.de} } @proceedings{DBLP:conf/icml/2003, editor = {T. Fawcett and N. Mishra}, title = {ICML}, publisher = {AAAI Press}, year = {2003}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/icml/Dietterich98, author = {T. Dietterich}, title = "The {MAXQ} Method for Hierarchical Reinforcement Learning", booktitle = {ICML}, year = {1998}, pages = {118-126}, crossref = {DBLP:conf/icml/1998}, bibsource = {DBLP, http://dblp.uni-trier.de} } @proceedings{DBLP:conf/icml/1998, editor = {Jude W. Shavlik}, title = {Proceedings of the Fifteenth International Conference on Machine Learning (ICML 1998), Madison, Wisconson, USA, July 24-27, 1998}, booktitle = {ICML}, publisher = {Morgan Kaufmann}, year = {1998}, isbn = {1-55860-556-8}, bibsource = {DBLP, http://dblp.uni-trier.de} } @phdthesis{ parr98hierarchical, author = "R. Parr", title = "Hierarchical Control and learning for {Markov} decision processes", text = "Parr, R. (1998). Hierarchical Control and learning for Markov decision processes. PhD thesis, University of California at Berkeley.", year = "1998", school = "University of California at Berkeley", url = "citeseer.ist.psu.edu/parr98hierarchical.html" } @inproceedings{Pineau_2001_4830, author = "J. Pineau and N. Roy and S. Thrun", title = "A Hierarchical Approach to {POMDP} Planning and Execution", booktitle = "Workshop on Hierarchy and Memory in Reinforcement Learning (ICML)", year = "2001" } @phdthesis{Pineau_2004_4767, author = "J. Pineau", title = "Tractable Planning Under Uncertainty: Exploiting Structure", school = "Robotics Institute, Carnegie Mellon University", year = "2004" } @phdthesis{cassandra:phd98, author = "A. Cassandra", title = "Exact and approximate algorithms for partially observable {Markov} decision processes", school = "Brown University, Dept. of Computer Science", year = "1998" } @techreport{MIT-CSAIL-TR-2005-058, author = {G. Theocharous and S. Mahadevan and L. Kaelbling}, title = "Spatial and Temporal Abstractions in {POMDPs} Applied to Robot Navigation", institution = {Computer Science and Artificial Intelligence Laboratory, MIT}, number = {MIT-CSAIL-TR-2005-058}, year = {2005} } @article{DBLP:journals/jair/Dietterich00, author = {T. Dietterich}, title = "Hierarchical Reinforcement Learning with the {MAXQ} Value Function Decomposition", journal = {JAIR}, volume = {13}, year = {2000}, pages = {227-303}, ee = {http://www.cs.washington.edu/research/jair/abstracts/dietterich00a.html}, bibsource = {DBLP, http://dblp.uni-trier.de} } @article{DBLP:journals/ai/SuttonPS99, author = {R. Sutton and D. Precup and S. Singh}, title = "Between {MDPs} and {Semi-MDPs}: A Framework for Temporal Abstraction in Reinforcement Learning", journal = {Artificial Intelligence}, volume = {112}, number = {1-2}, year = {1999}, pages = {181-211}, ee = {http://dx.doi.org/10.1016/S0004-3702(99)00052-1}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/uai/PineauGT03, author = {J. Pineau and G. Gordon and S. Thrun}, title = {Policy-contingent abstraction for robust robot control.}, booktitle = {UAI}, year = {2003}, pages = {477-484}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/nips/ThrunS94, author = {S. Thrun and A. Schwartz}, title = {Finding Structure in Reinforcement Learning.}, booktitle = {NIPS}, year = {1994}, pages = {385-392}, ee = {http://nips.djvuzone.org/djvu/nips07/0385.djvu}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/icml/Hengst02, author = {B. Hengst}, title = "Discovering Hierarchy in Reinforcement Learning with {HEXQ}", booktitle = {ICML}, year = {2002}, pages = {243-250}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/icml/Ryan02, author = {M. Ryan}, title = {Using Abstract Models of Behaviours to Automatically Generate Reinforcement Learning Hierarchies.}, booktitle = {ICML}, year = {2002}, pages = {522-529}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{DBLP:conf/icml/McGovernB01, author = {A. McGovern and A. Barto}, title = {Automatic Discovery of Subgoals in Reinforcement Learning using Diverse Density.}, booktitle = {ICML}, year = {2001}, pages = {361-368}, bibsource = {DBLP, http://dblp.uni-trier.de} } @inproceedings{williams-young:aaai06, author = {J. Williams and S. Youngs}, title = "Scaling {POMDPs} for dialogue management with composite summary point-based value iteration {(CSPBVI)}", booktitle = {AAAI workshop on Statistical and Empirical Methods in Spoken Dialogue Systems}, year = {2006} } Jason D. Williams and Steve Young. 2006. Scaling POMDPs for dialog management with composite summary point-based value iteration (CSPBVI). To appear, Proc AAAI Workshop on statistical and empirical methods in spoken dialog systems, Boston, USA.