Structured weight matrices-Based hardware accelerators in deep neural networks: FPGAs and ASICs

Caiwen Ding, Ao Ren, Geng Yuan, Xiaolong Ma, Jiayu Li, Ning Liu, Bo Yuan, Yanzhi Wang

Research output: Chapter in Book/Entry/PoemConference contribution

14 Scopus citations


Both industry and academia have extensively investigated hardware accelerations. In this work, to address the increasing demands in computational capability and memory requirement, we propose structured weight matrices (SWM)-based compression techniques for both field programmable gate array (FPGA) and application-specific integrated circuit (ASIC) implementations. In algorithm part, SWM-based framework adopts block-circulant matrices to achieve a fine-grained tradeoff between accuracy and compression ratio. The SWM-based technique can reduce computational complexity from O(n2) to O(n log n) and storage complexity from O(n2) to O(n) for each layer and both training and inference phases. For FPGA implementations on deep convolutional neural networks (DCNNs), we achieve at least 152X and 72X improvement in performance and energy efficiency, respectively using the SWM-based framework, compared with the baseline of IBM TrueNorth processor under same accuracy constraints using the data set of MNIST, SVHN, and CIFAR-10. For FPGA implementations on long short term memory (LSTM) networks, the proposed SWM-based LSTM can achieve up to 21X enhancement in performance and 33.5X gains in energy efficiency compared with the baseline accelerator. For ASIC implementations, the SWM-based ASIC design exhibits impressive advantages in terms of power, throughput, and energy efficiency. Experimental results indicate that this method is greatly suitable for applying DNNs onto both FPGAs and mobile/IoT devices.

Original languageEnglish (US)
Title of host publicationGLSVLSI 2018 - Proceedings of the 2018 Great Lakes Symposium on VLSI
PublisherAssociation for Computing Machinery
Number of pages6
ISBN (Electronic)9781450357241
StatePublished - May 30 2018
Event28th Great Lakes Symposium on VLSI, GLSVLSI 2018 - Chicago, United States
Duration: May 23 2018May 25 2018

Publication series

NameProceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI


Other28th Great Lakes Symposium on VLSI, GLSVLSI 2018
Country/TerritoryUnited States


  • ASIC
  • Accelerator
  • Deep learning
  • FPGA
  • Structured weight matrices

ASJC Scopus subject areas

  • General Engineering


Dive into the research topics of 'Structured weight matrices-Based hardware accelerators in deep neural networks: FPGAs and ASICs'. Together they form a unique fingerprint.

Cite this