Multifold Acceleration of Neural Network Computations Using GPU

  • Alexander Guzhva
  • Sergey Dolenko
  • Igor Persiantsev
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5768)


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).


GPGPU neural networks perceptron NVIDIA CUDA parallel computations 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    NVIDIA Corporation, CUDA Zone,
  2. 2.
    Kyoung-Su Oh, K.S., Jung, K.: GPU implementation of neural networks. Pattern Recognition 37, 1311–1314 (2004)CrossRefzbMATHGoogle Scholar
  3. 3.
    Zhongwen, L., Hongzhi, L., Zhengping, Y., Xincai, W.: Self-Organizing Maps Computing on Graphic Process Unit. In: ESANN 2005 proceedings, European Symposium on Artificial Neural Networks Bruges, Belgium, April 27-29. d-side publi. (2005) ISBN 2-930307-05-6,
  4. 4.
    Lahabar, S., Agrawal, P., Narayanan, P.J.: High Performance Pattern Recognition on GPU,
  5. 5.
    Guzhva, A.G., Dolenko, S.A., Obornev, E.A., Persiantsev, I.G., Shimelevich, M.I., Shugai, J.S.: Use of Significant Feature Selection Adaptive Algorithms in Neural Networks Based Solution of the Inverse Problem of Electrical Prospecting. In: 9th International Conference Pattern Recognition and Image Analysis: New Informational Technologies (PRIA 2008), Conference Proceedings, vol. 1, pp. 215–218. Nizhni Novgorod, Russia (2008)Google Scholar
  6. 6.
    NVIDIA Corporation,

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Alexander Guzhva
    • 1
  • Sergey Dolenko
    • 1
  • Igor Persiantsev
    • 1
  1. 1.D.V. Skobeltsyn Institute of Nuclear PhysicsM.V. Lomonosov Moscow State UniversityMoscowRussia

Personalised recommendations