Accelerating Convolutional Neural Network by Exploiting Sparsity on GPUs
ACM Transactions on Architecture and Code Optimization · 30.09.2023 · 20 (3) · 1-26
DOI: 10.1145/3600092 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 19.07.2023. Verfasst von Weizhi Xu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 96 Literaturangaben. Gefördert von: National Science Foundation, Natural Science Foundation of Shandong Province, National Natural Science Foundation of China.
Abstract (Originalsprache)
The convolutional neural network (CNN) is an important deep learning method, which is widely used in many fields. However, it is very time consuming to implement the CNN where convolution usually takes most of the time. There are many zero values in feature maps and filters, which leads to redundant calculations and memory accesses if dense methods are used to compute convolution. Many works recently have made use of sparsity to skip the calculations for zero values to reduce the inference time of the CNN. On the graphics processing unit platform, current works cannot fully exploit the sparsity of the feature map and achieve satisfactory performance. Therefore, we design a new parallel strategy to transform the feature map into a new storage format to avoid the redundant computation of zero values on graphics processing units. Also considering the sparsity in the feature map, we propose a fused storage format to combine the convolution operation with the following pooling operation, to further improve the performance. We carry out experiments with mainstream CNN models and achieve better performance compared with cuDNN and cuSPARSE. For VGG-19, ResNet-50, DenseNet-121, and RegNetX-16GF, 1.97×, 2.23×, 2.74×, and 1.58× speedups respectively are obtained over cuDNN. The speedups over cuSPARSE respectively are 2.10×, 1.83×, 2.35×, and 1.35× when only using the first method.
Förderung
- National Science Foundation
- Natural Science Foundation of Shandong Province
- National Natural Science Foundation of China
- Funding for Study Abroad Program by the Government of Shandong Province
Zitieren
APA
Xu, W., Sun, Y., Fan, S., Yu, H., & Fu, X. (2023). Accelerating Convolutional Neural Network by Exploiting Sparsity on GPUs. ACM Transactions on Architecture and Code Optimization, 20(3), 1-26. https://doi.org/10.1145/3600092
Vancouver
Xu W, Sun Y, Fan S, Yu H, Fu X. Accelerating Convolutional Neural Network by Exploiting Sparsity on GPUs. ACM Transactions on Architecture and Code Optimization. 2023;20(3):1-26. doi:10.1145/3600092
BibTeX
@article{Xu2023,
author = {Xu, Weizhi and Sun, Yintai and Fan, Shengyu and Yu, Hui and Fu, Xin},
title = {Accelerating Convolutional Neural Network by Exploiting Sparsity on GPUs},
journal = {ACM Transactions on Architecture and Code Optimization},
year = {2023},
volume = {20},
number = {3},
pages = {1--26},
doi = {10.1145/3600092}
}
EUVIMED