Reference : Approximate Policy Iteration for Closed-Loop Learning of Visual Tasks
Scientific congresses and symposiums : Paper published in a journal
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/1414
Approximate Policy Iteration for Closed-Loop Learning of Visual Tasks
English
Jodogne, Sébastien mailto [Centre Hospitalier Universitaire de Liège - CHU > > Radiothérapie >]
Briquet, Cyril mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Informatique (ingénierie du logiciel et algorithmique) >]
Piater, Justus mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > INTELSIG Group >]
Sep-2006
Lecture Notes in Computer Science
Springer
4212
210-221
Yes
No
International
0302-9743
1611-3349
Berlin
Germany
European Conference on Machine Learning (ECML)
du 18 septembre 2006 au 22 septembre 2006
Berlin
Germany
[en] Reinforcement Learning ; Approximate Policy Iteration ; Extra-Trees ; Grid computing ; BitTorrent
[en] Approximate Policy Iteration (API) is a reinforcement learning paradigm that is able to solve high- dimensional, continuous control problems. We propose to exploit API for the closed-loop learning of mappings from images to actions. This approach requires a family of function approximators that maps visual percepts to a real-valued function. For this purpose, we use Regression Extra-Trees, a fast, yet accurate and versatile machine learning algorithm. The inputs of the Extra-Trees consist of a set of visual features that digest the informative patterns in the visual signal. We also show how to parallelize the Extra-Tree learning process to further reduce the computational expense, which is often essential in visual tasks. Experimental results on real-world images are given that indicate that the combination of API with Extra-Trees is a promising framework for the interactive learning of visual tasks.
Researchers ; Professionals
http://hdl.handle.net/2268/1414
http://www.montefiore.ulg.ac.be/~jodogne/papers/JBP06.pdf
http://www.springerlink.com/content/u7541v105r6t7gv0/fulltext.pdf

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
JBP06.pdfThe original publication is available at www.springerlink.comAuthor postprint1.39 MBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.