Reference : Variable selection for dynamic treatment regimes: a reinforcement learning approach
Scientific congresses and symposiums : Unpublished conference
Engineering, computing & technology : Multidisciplinary, general & others
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/13599
Variable selection for dynamic treatment regimes: a reinforcement learning approach
English
Fonteneau, Raphaël 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 >]
2008
7
Yes
No
International
European Workshop on Reinforcement Learning 2008 (EWRL'08)
30 June - 3 July
Villeneuve d'Ascq
France
[en] dynamic treatment regimes ; reinforcement learning ; fitted Q iteration
[en] Dynamic treatment regimes (DTRs) can be inferred from data collected through some randomized clinical trials by using reinforcement learning algorithms. During these clinical trials, a large set of clinical indicators are usually monitored. However, it is often more convenient for clinicians to have DTRs which are only defined on a small set of indicators rather than on the original full set. To address this problem, we analyse the approximation architecture of the state-action value functions computed by the fitted Q iteration algorithm - a RL algorithm - using tree-based regressors in order to identify a small subset of relevant ones. The RL algorithm is then rerun by considering only as state variables these most relevant indicators to have DTRs defined on a small set of indicators. The approach is validated on benchmark problems inspired
from the classical ‘car on the hill’ problem and the results obtained are positive.
Fonds de la Recherche Scientifique (Communauté française de Belgique) - F.R.S.-FNRS
http://hdl.handle.net/2268/13599

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
raphael-vs.pdfPublisher postprint103.57 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.