Reference : Reinforcement learning with raw image pixels as input state
Scientific congresses and symposiums : Paper published in a book
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/1113
Reinforcement learning with raw image pixels as input state
English
Ernst, Damien mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Marée, Raphaël mailto [Université de Liège - ULg > > Systèmes et modélisation >]
Wehenkel, Louis mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
2006
Advances in machine vision, image processing & pattern analysis (Lecture notes in computer science, Vol. 4153)
Springer-Verlag Berlin
Yes
No
International
978-3-540-37597-5
Berlin
International Workshop on Intelligent Computing in Pattern Analysis/Synthesis (IWICPAS)
August 26-27, 2006
Xi'an
China
[en] Reinforcement Learning ; image ; fitted Q iteration
[en] We report in this paper some positive simulation results obtained when image pixels are directly used as input state of a reinforcement learning algorithm. The reinforcement learning algorithm chosen to carry out the simulation is a batch-mode algorithm known as fitted Q iteration.
Fonds de la Recherche Scientifique (Communauté française de Belgique) - F.R.S.-FNRS
http://hdl.handle.net/2268/1113
http://www.springerlink.com/content/978-3-540-37597-5/
http://www.montefiore.ulg.ac.be/~ernst/

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
ernst-iwicpas-2006.pdfNo commentaryAuthor postprint181.73 kBView/Open

Additional material(s):

File Commentary Size Access
Open access
ernst-iwicpas-2006-slides.pdfnull199.14 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.