Saturation RRAM leveraging bit-level sparsity resulting from term quantization

Bradley McDanel, H. T. Kung, Sai Qian Zhang

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

Abstract

The proposed saturation RRAM for in-memory computing of a pre-trained Convolutional Neural Network (CNN) inference imposes a limit on the maximum analog value output from each bitline in order to reduce analog-to-digital (A/D) conversion costs. The proposed scheme uses term quantization (TQ) to enable flexible bit annihilation at any position for a value in the context of a group of weights values in RRAM. This enables a drastic reduction in the required ADC resolution while still maintaining CNN model accuracy. Specifically, we show that the A/D conversion errors after TQ have a minimum impact on the classification accuracy of the inference task. For instance, for a 64x64 RRAM, reducing the ADC resolution from 6 bits to 4 bits enables a 1.58x reduction in the total system power, without a significant impact to classification accuracy.

Original languageEnglish (US)
Title of host publication2021 IEEE International Symposium on Circuits and Systems, ISCAS 2021 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728192017
DOIs
StatePublished - 2021
Event53rd IEEE International Symposium on Circuits and Systems, ISCAS 2021 - Daegu, Korea, Republic of
Duration: May 22 2021May 28 2021

Publication series

NameProceedings - IEEE International Symposium on Circuits and Systems
Volume2021-May
ISSN (Print)0271-4310

Conference

Conference53rd IEEE International Symposium on Circuits and Systems, ISCAS 2021
Country/TerritoryKorea, Republic of
CityDaegu
Period5/22/215/28/21

Keywords

  • Analog computing
  • Analog-to-digital conversion (A/D conversion)
  • Analog-to-digital converter (ADC)
  • Convolutional neural network (CNN)
  • Dot-product computation
  • In-memory computing
  • Noise
  • Resistive RAM (RRAM)

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Saturation RRAM leveraging bit-level sparsity resulting from term quantization'. Together they form a unique fingerprint.

Cite this