Unpublished conference/Abstract (Scientific congresses and symposiums)
Large Margin Classification with the Progressive Hedging Algorithm
Defourny, Boris; Wehenkel, Louis
2009Second NIPS Workshop on Optimization for Machine Learning
 

Files


Full Text
OPT2009-Defourny.pdf
Author postprint (112.98 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
SVM; parallel algorithms; stochastic programming
Abstract :
[en] Several learning algorithms in classification and structured prediction are formulated as large scale optimization problems. We show that a generic iterative reformulation and resolving strategy based on the progressive hedging algorithm from stochastic programming results in a highly parallel algorithm when applied to the large margin classification problem with nonlinear kernels. We also underline promising aspects of the available analysis of progressive hedging strategies.
Research Center/Unit :
Systems and Modeling Research Unit
Disciplines :
Computer science
Author, co-author :
Defourny, Boris ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Dép. d'électric., électron. et informat. (Inst.Montefiore)
Wehenkel, Louis  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Language :
English
Title :
Large Margin Classification with the Progressive Hedging Algorithm
Publication date :
December 2009
Event name :
Second NIPS Workshop on Optimization for Machine Learning
Event organizer :
Sebastian Nowozin; Suvrit Sra; SVN Vishwanathan; Stephen Wright
Event place :
Whistler, Canada
Event date :
December 12, 2009
Audience :
International
Funders :
DYSCO (Dynamical Systems, Control, and Optimization); PASCAL2 Network of Excellence
Available on ORBi :
since 26 December 2010

Statistics


Number of views
118 (4 by ULiège)
Number of downloads
85 (2 by ULiège)

Bibliography


Similar publications



Sorry the service is unavailable at the moment. Please try again later.
Contact ORBi