Proceedings of the 2017 International Conference Advanced Engineering and Technology Research (AETR 2017)

Research and Design of Neural Network Based on GPU

Authors
Jiaohua Yang, Xin Wang
Corresponding Author
Jiaohua Yang
Available Online March 2018.
DOI
10.2991/aetr-17.2018.15How to use a DOI?
Keywords
GPU; Artificial neural network; Graphics processor
Abstract

With the rapid development of graphics processor (GPU) programmable ability, coupled with its high speed and parallelism for large scale neural network BP algorithm and the problem of low efficiency, people put forward a kind of neural network BP algorithm based on GPU acceleration. Through in-depth study of the CUDA technology programming and programming model to solve the complicated problem of parallel computing using this framework, converse the process of the BP neural network in CPU forward calculation and reverse learning to the process of learning accelerated in the GPU, and then use the GPU powerful floating-point computation ability and high parallel computing characteristics to achieve BP algorithm. Finally people design and achieve a kind of neural network based on the GPU acceleration training.

Copyright
© 2018, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

Volume Title
Proceedings of the 2017 International Conference Advanced Engineering and Technology Research (AETR 2017)
Series
Advances in Engineering Research
Publication Date
March 2018
ISBN
978-94-6252-489-7
ISSN
2352-5401
DOI
10.2991/aetr-17.2018.15How to use a DOI?
Copyright
© 2018, the Authors. Published by Atlantis Press.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - CONF
AU  - Jiaohua Yang
AU  - Xin Wang
PY  - 2018/03
DA  - 2018/03
TI  - Research and Design of Neural Network Based on GPU
BT  - Proceedings of the 2017 International Conference Advanced Engineering and Technology Research (AETR 2017)
PB  - Atlantis Press
SP  - 69
EP  - 72
SN  - 2352-5401
UR  - https://doi.org/10.2991/aetr-17.2018.15
DO  - 10.2991/aetr-17.2018.15
ID  - Yang2018/03
ER  -