Multifold acceleration of neural network computations using GPUстатья
Информация о цитировании статьи получена из
Web of Science,
Scopus
Дата последнего поиска статьи во внешних источниках: 19 июля 2013 г.
-
Авторы:
Guzhva A.,
Dolenko S.,
Persiantsev I.
-
Сборник:
International Conference on Artificial Neural Networks (ICANN-2009), Part 1
-
Серия:
Lecture Notes in Computer Science
-
Том:
5768
-
Год издания:
2009
-
Место издания:
Springer-Verlag Berlin Heidelberg
-
Первая страница:
373
-
Последняя страница:
380
-
DOI:
10.1007/978-3-642-04274-4_39
-
Аннотация:
With emergence of graphics processing units (GPU) of the latest generation, it became possible to undertake neural network based computations using GPU on serially produced video display adapters. In this study, NVIDIA CUDA technology has been used to implement standard back-propagation algorithm for training multiple perceptrons simultaneously on GPU. For the problem considered, GPU-based implementation (on NVIDIA GTX 260 GPU) has lead to a 50x speed increase compared to a highly optimized CPU-based computer program, and more than 150x compared to a commercially available CPU-based software (NeuroShell 2) (AMD Athlon 64 Dual core 6000+ processor). В© 2009 Springer Berlin Heidelberg.
-
Добавил в систему:
Доленко Сергей Анатольевич