TY - GEN
T1 - Defending bit-flip attack through DNN weight reconstruction
AU - Li, Jingtao
AU - Rakin, Adnan Siraj
AU - Xiong, Yan
AU - Chang, Liangliang
AU - He, Zhezhi
AU - Fan, Deliang
AU - Chakrabarti, Chaitali
N1 - Publisher Copyright:
© 2020 IEEE.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/7
Y1 - 2020/7
N2 - Recent studies show that adversarial attacks on neural network weights, aka, Bit-Flip Attack (BFA), can degrade Deep Neural Network's (DNN) prediction accuracy severely. In this work, we propose a novel weight reconstruction method as a countermeasure to such BFAs. Specifically, during inference, the weights are reconstructed such that the weight perturbation due to BFA is minimized or diffused to the neighboring weights. We have successfully demonstrated that our method can significantly improve the DNN robustness against random and gradient-based BFA variants. Even under the most aggressive attacks (i.e., greedy progressive bit search), our method maintains a test accuracy of 60% on ImageNet after 5 iterations while the baseline accuracy drops to below 1%.
AB - Recent studies show that adversarial attacks on neural network weights, aka, Bit-Flip Attack (BFA), can degrade Deep Neural Network's (DNN) prediction accuracy severely. In this work, we propose a novel weight reconstruction method as a countermeasure to such BFAs. Specifically, during inference, the weights are reconstructed such that the weight perturbation due to BFA is minimized or diffused to the neighboring weights. We have successfully demonstrated that our method can significantly improve the DNN robustness against random and gradient-based BFA variants. Even under the most aggressive attacks (i.e., greedy progressive bit search), our method maintains a test accuracy of 60% on ImageNet after 5 iterations while the baseline accuracy drops to below 1%.
KW - Bit-Flip Attack
KW - Row-Hammer Attack
KW - Security of Deep Neural Network
UR - http://www.scopus.com/inward/record.url?scp=85093928301&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85093928301&partnerID=8YFLogxK
U2 - 10.1109/DAC18072.2020.9218665
DO - 10.1109/DAC18072.2020.9218665
M3 - Conference contribution
AN - SCOPUS:85093928301
T3 - Proceedings - Design Automation Conference
BT - 2020 57th ACM/IEEE Design Automation Conference, DAC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 57th ACM/IEEE Design Automation Conference, DAC 2020
Y2 - 20 July 2020 through 24 July 2020
ER -