Carregant...
Carregant...

Vés al contingut (premeu Retorn)

Distributed training strategies for a computer vision deep learning algorithm on a distributed GPU cluster

Autor
Campos, V.; Sastre, F.; Yagües, M.; Bellver, M.; Giro, X.; Torres, J.
Tipus d'activitat
Article en revista
Revista
Procedia computer science
Data de publicació
2017
Volum
108
Pàgina inicial
315
Pàgina final
324
DOI
https://doi.org/10.1016/j.procs.2017.05.074 Obrir en finestra nova
Projecte finançador
Procesado de información heterogénea y señales en grafos para Big Data:aplicación en cribado de alto rendimiento,teledetección,multimedia y HCI
Repositori
http://hdl.handle.net/2117/107590 Obrir en finestra nova
URL
http://www.sciencedirect.com/science/article/pii/S1877050917306129 Obrir en finestra nova
Resum
Deep learning algorithms base their success on building high learning capacity models with millions of parameters that are tuned in a data-driven fashion. These models are trained by processing millions of examples, so that the development of more accurate algorithms is usually limited by the throughput of the computing devices on which they are trained. In this work, we explore how the training of a state-of-the-art neural network for computer vision can be parallelized on a distributed GPU clu...
Citació
Campos, V., Sastre, F., Yagües, M., Bellver, M., Giro, X., Torres, J. Distributed training strategies for a computer vision deep learning algorithm on a distributed GPU cluster. "Procedia computer science", 2017, vol. 108, p. 315-324.
Paraules clau
Convolutional Neural Networks, Deep learning, Distributed computing, Parallel systems
Grup de recerca
CAP - Grup de Computació d'Altes Prestacions
GPI - Grup de Processament d'Imatge i Vídeo
IDEAI-UPC Intelligent Data Science and Artificial Intelligence

Participants