Abstract
The main contribution of this paper is to show efficient implementations of the convolution-pooling in the GPU, in which the pooling follows the multiple convolution. Since the multiple convolution and the pooling operations are performed alternately in earlier stages of many Convolutional Neural Networks (CNNs), it is very important to accelerate the convolution-pooling. Our new GPU implementation uses two techniques, (1) convolution interchange with direct sum, and (2) conversion to matrix multiplication. By these techniques, the computational and memory access cost are reduced. Further the convolution interchange is converted to matrix multiplication, which can be computed by cuBLAS very efficiently. Experimental results using Telsa V100 GPU show that our new GPU implementation compatible with cuDNN for the convolution-pooling is at least 1.34 times faster than the multiple convolution and then the pooling by cuDNN, the most popular library of primitives to implement the CNNs in the GPU.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Cheng, Y., Wang, D., Zhou, P., Zhang, T.: A survey of model compression and acceleration for deep neural networks. CoRR abs/1710.09282, October 2017
Chetlur, S., Woolley, C., Vandermersch, P., Cohen, J., Tran, J., Catanzaro, B., Shelhamer, E.: cuDNN: efficient primitives for deep learning. CoRR abs/1410.0759, August 2014
Emoto, Y., Funasaka, S., Tokura, H., Honda, T., Nakano, K., Ito, Y.: An optimal parallel algorithm for computing the summed area table on the GPU. In: Proceedings of International Parallel and Distributed Processing Symposium Workshops, pp. 763–772, February 2018
Honda, T., Yamamoto, S., Honda, H., Nakano, K., Ito, Y.: Simple and fast parallel algorithms for the Voronoi map and the Euclidean distance map, with GPU implementations. In: Proceedings of International Conference on Parallel Processing, pp. 362–371, August 2017
Hwu, W.W.: GPU Computing Gems, Emerald edn. Morgan Kaufmann, Burlington (2011)
Kasagi, A., Nakano, K., Ito, Y.: Parallel algorithms for the summed area table on the asynchronous hierarchical memory machine, with GPU implementations. In: Proceedings of International Conference on Parallel Processing (ICPP), pp. 251–260, September 2014
Kasagi, A., Tabaru, T., Tamura, H.: Fast algorithm using summed area tables with unified layer performing convolution and average pooling. In: Proceedings of International Workshop on Machine Learning for Signal Processing, September 2017
Li, C., Yang, Y., Feng, M., Chakradhar, S., Zhou, H.: Optimizing memory efficiency for deep convolutional neural networks on GPUs. In: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, November 2016
Matsumura, N., Tokura, H., Kuroda, Y., Ito, Y., Nakano, K.: Tile art image generation using conditional generative adversarial networks. In: Proceedings of International Symposium on Computing and Networking Workshops, pp. 209–215 (2018)
NVIDIA Corporation: NVIDIA CUDA C programming guide version 4.0 (2011)
NVIDIA Corporation: CUBLAS LIBRARY user guide, February 2019. https://docs.nvidia.com/cuda/cublas/index.html
NVIDIA Corporation: CUDNN developer guide, February 2019. https://docs.nvidia.com/deeplearning/sdk/cudnn-developer-guide/index.html
Ogawa, K., Ito, Y., Nakano, K.: Efficient Canny edge detection using a GPU. In: Proceedings of International Conference on Networking and Computing, pp. 279–280. IEEE CS Press, November 2010
Sze, V., Chen, Y.H., Yang, T.J., Emer, J.S.: Efficient processing of deep neural networks: a tutorial and survey. Proc. IEEE 105(12), 2295–2329 (2017)
Takeuchi, Y., Takafuji, D., Ito, Y., Nakano, K.: ASCII art generation using the local exhaustive search on the GPU. In: Proceedings of International Symposium on Computing and Networking, pp. 194–200, December 2013
Zhang, Q., Zhang, M., Chen, T., Sun, Z., Ma, Y., Yu, B.: Recent advances in convolutional neural network acceleration. Neurocomputing 323, 37–51 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Suita, S. et al. (2020). Efficient cuDNN-Compatible Convolution-Pooling on the GPU. In: Wyrzykowski, R., Deelman, E., Dongarra, J., Karczewski, K. (eds) Parallel Processing and Applied Mathematics. PPAM 2019. Lecture Notes in Computer Science(), vol 12044. Springer, Cham. https://doi.org/10.1007/978-3-030-43222-5_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-43222-5_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-43221-8
Online ISBN: 978-3-030-43222-5
eBook Packages: Computer ScienceComputer Science (R0)