Mapping Binary ResNets on Computing-In-Memory Hardware with Low-bit ADCs

Yulhwa Kim, Hyungjun Kim, Jihoon Park, Hyunmyung Oh, Jae Joon Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations

Abstract

Implementing binary neural networks (BNNs) on computing-in-memory (CIM) hardware has several attractive features such as small memory requirement and minimal overhead in peripheral circuits such as analog-to-digital converters (ADCs). On the other hand, one of the downsides of using BNNs is that it degrades the classification accuracy. Recently, ResNet-style BNNs are gaining popularity with higher accuracy than conventional BNNs. The accuracy improvement comes from the high-resolution skip connection which binary ResNets use to compensate the information loss caused by binarization. However, the high-resolution skip connection forces the CIM hardware to use high-bit ADCs again so that area and energy overhead becomes larger. In this paper, we demonstrate that binary ResNets can be also mapped on CIM with low-bit ADCs via aggressive partial sum quantization and input-splitting combined with retraining. As a result, the key advantages of BNN CIM such as small area and energy consumption can be preserved with higher accuracy.

Original languageEnglish
Title of host publicationProceedings of the 2021 Design, Automation and Test in Europe, DATE 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages856-861
Number of pages6
ISBN (Electronic)9783981926354
DOIs
StatePublished - 1 Feb 2021
Externally publishedYes
Event2021 Design, Automation and Test in Europe Conference and Exhibition, DATE 2021 - Virtual, Online
Duration: 1 Feb 20215 Feb 2021

Publication series

NameProceedings -Design, Automation and Test in Europe, DATE
Volume2021-February
ISSN (Print)1530-1591

Conference

Conference2021 Design, Automation and Test in Europe Conference and Exhibition, DATE 2021
CityVirtual, Online
Period1/02/215/02/21

Keywords

  • analog computing
  • computing in memory
  • hardware-Nn co-design
  • NN accelerator

Fingerprint

Dive into the research topics of 'Mapping Binary ResNets on Computing-In-Memory Hardware with Low-bit ADCs'. Together they form a unique fingerprint.

Cite this