CorrectNet: Robustness Enhancement of Analog In-Memory Computing for Neural Networks by Error Suppression and Compensation

Amro Eldebiky, Grace Li Zhang, Georg Böcherer, Bing Li, Ulf Schlichtmann

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

8 Scopus citations

Abstract

The last decade has witnessed the breakthrough of deep neural networks (DNNs) in many fields. With the increasing depth of DNNs, hundreds of millions of multiply-and-accumulate (MAC) operations need to be executed. To accelerate such operations efficiently, analog in-memory computing platforms based on emerging devices, e.g., resistive RAM (RRAM), have been introduced. These acceleration platforms rely on analog properties of the devices and thus suffer from process variations and noise. Consequently, weights in neural networks configured into these platforms can deviate from the expected values, which may lead to feature errors and a significant degradation of inference accuracy. To address this issue, in this paper, we propose a framework to enhance the robustness of neural networks under variations and noise. First, a modified Lipschitz constant regularization is proposed during neural network training to suppress the amplification of errors propagated through network layers. Afterwards, error compensation is introduced at necessary locations determined by reinforcement learning to rescue the feature maps with remaining errors. Experimental results demonstrate that inference accuracy of neural networks can be recovered from as low as 1.69% under variations and noise back to more than 95% of their original accuracy, while the training and hardware cost are negligible.

Original languageEnglish
Title of host publication2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9783981926378
DOIs
StatePublished - 2023
Event2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023 - Antwerp, Belgium
Duration: 17 Apr 202319 Apr 2023

Publication series

NameProceedings -Design, Automation and Test in Europe, DATE
Volume2023-April
ISSN (Print)1530-1591

Conference

Conference2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023
Country/TerritoryBelgium
CityAntwerp
Period17/04/2319/04/23

Fingerprint

Dive into the research topics of 'CorrectNet: Robustness Enhancement of Analog In-Memory Computing for Neural Networks by Error Suppression and Compensation'. Together they form a unique fingerprint.

Cite this