Scholars@Duke publication: An FPGA design framework for CNN sparsification and acceleration

An FPGA design framework for CNN sparsification and acceleration

Publication , Conference

Li, S; Wen, W; Wang, Y; Han, S; Chen, Y; Li, HH

Published in: Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017

June 30, 2017

Published version (DOI)

Convolutional neural networks (CNNs) have recently broken many performance records in image recognition and object detection problems. The success of CNNs, to a great extent, is enabled by the fast scaling-up of the networks that learn from a huge volume of data. The deployment of big CNN models can be both computation-intensive and memory-intensive, leaving severe challenges to hardware implementations. In recent years, sparsification techniques that prune redundant connections in the networks while still retaining the similar accuracy emerge as promising solutions to alliterate the computation overheads associated with CNNs [1].

Duke Scholars

Author Yiran Chen Electrical and Computer Engineering

Author Hai "Helen" Li Electrical and Computer Engineering

Published In

Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017

DOI

10.1109/FCCM.2017.21

Publication Date

June 30, 2017

Start / End Page

Citation

APA

Chicago

ICMJE

MLA

NLM

Li, S., Wen, W., Wang, Y., Han, S., Chen, Y., & Li, H. H. (2017). An FPGA design framework for CNN sparsification and acceleration. In Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017 (p. 28). https://doi.org/10.1109/FCCM.2017.21

Li, S., W. Wen, Y. Wang, S. Han, Y. Chen, and H. H. Li. “An FPGA design framework for CNN sparsification and acceleration.” In Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017, 28, 2017. https://doi.org/10.1109/FCCM.2017.21.

Li S, Wen W, Wang Y, Han S, Chen Y, Li HH. An FPGA design framework for CNN sparsification and acceleration. In: Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017. 2017. p. 28.

Li, S., et al. “An FPGA design framework for CNN sparsification and acceleration.” Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017, 2017, p. 28. Scopus, doi:10.1109/FCCM.2017.21.

Li S, Wen W, Wang Y, Han S, Chen Y, Li HH. An FPGA design framework for CNN sparsification and acceleration. Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017. 2017. p. 28.

Published In

Proceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017