Unpublished conference/Abstract (Scientific congresses and symposiums)
Tree based ensemble models regularization by convex optimization
Cornélusse, Bertrand; Geurts, Pierre; Wehenkel, Louis
2009NIPS-09 workshop on Optimization for Machine Learning
 

Files


Full Text
OPT2009_f.pdf
Author preprint (136.61 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Ensemble methods; regularization; Convex optimization
Abstract :
[en] Tree based ensemble methods can be seen as a way to learn a kernel from a sample of input-output pairs. This paper proposes a regularization framework to incorporate non-standard information not used in the kernel learning algorithm, so as to take advantage of incomplete information about output values and/or of some prior information about the problem at hand. To this end a generic convex optimization problem is formulated which is first customized into a manifold regularization approach for semi-supervised learning, then as a way to exploit censored output values, and finally as a generic way to exploit prior information about the problem.
Research center :
SystMod
Disciplines :
Electrical & electronics engineering
Author, co-author :
Cornélusse, Bertrand  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Geurts, Pierre ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Wehenkel, Louis  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Language :
English
Title :
Tree based ensemble models regularization by convex optimization
Publication date :
12 December 2009
Event name :
NIPS-09 workshop on Optimization for Machine Learning
Event place :
Whistler, Canada
Event date :
12-12-2009
Audience :
International
Funders :
FRIA - Fonds pour la Formation à la Recherche dans l'Industrie et dans l'Agriculture [BE]
Available on ORBi :
since 23 November 2009

Statistics


Number of views
202 (47 by ULiège)
Number of downloads
118 (22 by ULiège)

Bibliography


Similar publications



Contact ORBi