Abstract
In recent years, deep learning represented by Convolutional Neural Network (CNN) has been one of the hottest topics of research. CNN inference process based models have been widely used in more and more computer vision applications. The execution speed of inference process is critical for applications, and the hardware acceleration method is mostly considered. To relieve the memory pressure, data quantization strategies are often used in hardware implementation. In this paper, a dynamic multi-precision fixed-point data quantization strategy for CNN has been proposed and used to quantify the floating-point data in trained CNN inference process. Results shows that our quantization strategy for LeNet model can reduce the accuracy loss from 22.2% to 5.9% at most, compared with previous static quantization strategy, when 8/4-bit quantization is used. When 16-bit quantization is used, only 0.03% accuracy loss is introduced by our quantization strategy with half memory footprint and bandwidth requirement comparing with 32-bit floating-point implementation.
L. Shan—The research is supported by Specialized Research Fund for the Doctor Program of Higher Education of China with Grant No. 20124307110016, and by National Natural Science Foundation of China with Grant No. 61176030.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A.C., Fei-Fei, L.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115, 211–252 (2015)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NIPS, pp. 1097–1105 (2012)
Zeiler, M.D., Fergus, R.: Visualizing and understanding convolutional networks. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part I. LNCS, vol. 8689, pp. 818–833. Springer, Heidelberg (2014)
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. arXiv preprint arXiv:1409.4842 (2014)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. arXiv preprint arXiv:1512.03385 (2015)
Vanhoucke, V., Mao, M.Z.: Improving the speed of neural networks on CPUs. In: Deep Learning and Unsupervised Feature Learning Workshop NIPS (2011)
Chen, T., Du, Z., Sun, N., Wang, J., Wu, C., Chen, Y., Temam, O.: Diannao: a small-footprint high-throughput accelerator for ubiquitous machine-learning. In: ASPLOS, vol. 49, no. 4. ACM, pp. 269–284 (2014)
Chen, Y., Luo, T., Liu, S., Zhang, S., He, L., Wang, J., Li, L., Chen, T., Xu, Z., Sun, N., et al.: Dadiannao: a machine-learning supercomputer. In: MICRO. IEEE, pp. 609–622 (2014)
Qiu, J., Wang, J., Yao, S., et al.: Going deeper with embedded FPGA platform for convolutional neural network. In: ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. ACM (2016)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Shan, L., Zhang, M., Deng, L., Gong, G. (2016). A Dynamic Multi-precision Fixed-Point Data Quantization Strategy for Convolutional Neural Network. In: Xu, W., Xiao, L., Li, J., Zhang, C., Zhu, Z. (eds) Computer Engineering and Technology. NCCET 2016. Communications in Computer and Information Science, vol 666. Springer, Singapore. https://doi.org/10.1007/978-981-10-3159-5_10
Download citation
DOI: https://doi.org/10.1007/978-981-10-3159-5_10
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-3158-8
Online ISBN: 978-981-10-3159-5
eBook Packages: Computer ScienceComputer Science (R0)
