Leveraging Noise and Aggressive Quantization of In-Memory Computing for Robust DNN Hardware against Adversarial Input and Weight Attacks

Sai Kiran Cherupally, Adnan Siraj Rakin, Shihui Yin, Mingoo Seok, Deliang Fan, Jae Sun Seo

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

Abstract

In-memory computing (IMC) substantially improves the energy efficiency of deep neural network (DNNs) hardware by activating many rows together and performing analog computing. The noisy analog IMC induces some amount of accuracy drop in hardware acceleration, which is generally considered as a negative effect. However, in this work, we discover that such hardware intrinsic noise can, on the contrary, play a positive role in enhancing adversarial robustness. To achieve that, we propose a new DNN training scheme that integrates measured IMC hardware noise and aggressive partial sum quantization at the IMC crossbar. We show that this effectively improves the robustness of IMC DNN hardware against both adversarial input and weight attacks. Against black-box adversarial input attacks and bit-flip weight attacks, DNN robustness has improved by up to 10.5% (CFAR-10 accuracy) and 33.6% (number of bit-flips), respectively, compared to conventional DNNs.

Original languageEnglish (US)
Title of host publication2021 58th ACM/IEEE Design Automation Conference, DAC 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages559-564
Number of pages6
ISBN (Electronic)9781665432740
DOIs
StatePublished - Dec 5 2021
Event58th ACM/IEEE Design Automation Conference, DAC 2021 - San Francisco, United States
Duration: Dec 5 2021Dec 9 2021

Publication series

NameProceedings - Design Automation Conference
Volume2021-December
ISSN (Print)0738-100X

Conference

Conference58th ACM/IEEE Design Automation Conference, DAC 2021
Country/TerritoryUnited States
CitySan Francisco
Period12/5/2112/9/21

Keywords

  • adversarial attack
  • adversarial robustness
  • in-memory computing
  • low-precision quantization
  • noise injection

ASJC Scopus subject areas

  • Computer Science Applications
  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Modeling and Simulation

Fingerprint

Dive into the research topics of 'Leveraging Noise and Aggressive Quantization of In-Memory Computing for Robust DNN Hardware against Adversarial Input and Weight Attacks'. Together they form a unique fingerprint.

Cite this