DIMA: A Depthwise CNN In-Memory Accelerator

Shaahin Angizi, Zhezhi He, Deliang Fan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Scopus citations


In this work, we first propose a deep depthwise Convolutional Neural Network (CNN) structure, called Add-Net, which uses binarized depthwise separable convolution to replace conventional spatial-convolution. In Add-Net, the computationally expensive convolution operations (i.e. Multiplication and Accumulation) are converted into hardware-friendly Addition operations. We meticulously investigate and analyze the Add-Net's performance (i.e. accuracy, parameter size and computational cost) in object recognition application compared to traditional baseline CNN using the most popular large scale ImageNet dataset. Accordingly, we propose a Depthwise CNN In-Memory Accelerator (DIMA) based on SOT-MRAM computational sub-arrays to efficiently accelerate Add-Net within non-volatile MRAM. Our device-to-architecture co-simulation results show that, with almost the same inference accuracy to the baseline CNN on different data-sets, DIMA can obtain ∼1.4× better energy-efficiency and 15.7× speedup compared to ASICs, and, ∼1.6× better energy-efficiency and 5.6× speedup over the best processing-in-DRAM accelerators.

Original languageEnglish (US)
Title of host publication2018 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018 - Digest of Technical Papers
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781450359504
StatePublished - Nov 5 2018
Externally publishedYes
Event37th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018 - San Diego, United States
Duration: Nov 5 2018Nov 8 2018

Publication series

NameIEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
ISSN (Print)1092-3152


Conference37th IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2018
Country/TerritoryUnited States
CitySan Diego

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design


Dive into the research topics of 'DIMA: A Depthwise CNN In-Memory Accelerator'. Together they form a unique fingerprint.

Cite this