Reference : A zealous parallel gradient descent algorithm
Scientific congresses and symposiums : Poster
Engineering, computing & technology : Computer science
http://hdl.handle.net/2268/80780
A zealous parallel gradient descent algorithm
English
Louppe, Gilles mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
Geurts, Pierre mailto [Université de Liège - ULg > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation >]
11-Dec-2010
Yes
No
International
NIPS 2010 Workshop on Learning on Cores, Clusters and Clouds
11 décembre 2010
Whistler
Canada
[en] machine learning ; optimization ; gradient descent
[en] Parallel and distributed algorithms have become a necessity in modern machine
learning tasks. In this work, we focus on parallel asynchronous gradient descent and propose a zealous variant that minimizes the idle time of processors to achieve a substantial speedup. We then experimentally study this algorithm in the context of training a restricted Boltzmann machine on a large collaborative filtering task.
Fonds de la Recherche Scientifique (Communauté française de Belgique) - F.R.S.-FNRS
Researchers
http://hdl.handle.net/2268/80780

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
glouppe10-lccc-poster.pdfPosterAuthor preprint897.87 kBView/Open
Open access
glouppe-pgeurts-lccc2010.pdfExtended abstractAuthor preprint102.35 kBView/Open

Additional material(s):

File Commentary Size Access
Open access
rbm-zealous.zipSource code35.55 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.