Reference : Optimal sample selection for batch-mode reinforcement learning
Scientific congresses and symposiums : Paper published in a book
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/83529
Optimal sample selection for batch-mode reinforcement learning
English
Rachelson, Emmanuel mailto [> >]
Schnitzler, François mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Wehenkel, Louis mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Ernst, Damien mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
2011
Proceedings of the 3rd International Conference on Agents and Artificial Intelligence (ICAART 2011)
Yes
No
International
3rd International Conference on Agents and Artificial Intelligence (ICAART 2011)
28-30 January 2011
Rome
Italy
[en] stochastic optimal control ; sample control ; reinforcement learning
[en] We introduce the Optimal Sample Selection (OSS) meta-algorithm for solving discrete-time Optimal Control problems. This meta-algorithm maps the problem of finding a near-optimal closed-loop policy to the identification of a small set of one-step system transitions, leading to high-quality policies when used as input of a batch-mode Reinforcement Learning (RL) algorithm. We detail a particular instance of this OSS metaalgorithm that uses tree-based Fitted Q-Iteration as a batch-mode RL algorithm and Cross Entropy search as a method for navigating efficiently in the space of sample sets. The results show that this particular instance of OSS algorithms is able to identify rapidly small sample sets leading to high-quality policies
Fonds de la Recherche Scientifique (Communauté française de Belgique) - F.R.S.-FNRS
Researchers ; Professionals ; Students
http://hdl.handle.net/2268/83529

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
icaart-2011.pdfPublisher postprint775.65 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.