Parallel support vector machines on multi-core and multiprocessor systems - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2011

Parallel support vector machines on multi-core and multiprocessor systems

Résumé

This paper proposes a new and efficient parallel implementation of support vector machines based on decom- position method for handling large scale datasets. The paral- lelizing is performed on the most time-and-memory consuming work of training, i.e., to update the vector f . The inner problems are dealt by sequential minimal optimization solver. Since the underlying parallelism is realized by the shared memory version of Map-Reduce paradigm, our system is easy to build and particularly suitable to apply to multi-core and multiprocessor systems. Experimental results show that on most of the tested datasets, our system offers higher than four- fold increase in speed compared to Libsvm, and it is also far more efficient than the MPI implementation Pisvm.
Fichier principal
Vignette du fichier
MRPsvm.pdf (149.2 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00617933 , version 1 (31-08-2011)

Identifiants

Citer

H. X. Zhao, F. Magoules. Parallel support vector machines on multi-core and multiprocessor systems. 11th International Conference on Artificial Intelligence and Applications (AIA 2011), Feb 2011, Innsbruck, Austria. ⟨10.2316/P.2011.717-056⟩. ⟨hal-00617933⟩
206 Consultations
1887 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More