AccHashtag : Accelerated Hashing for Detecting Fault-Injection Attacks on Embedded Neural Networks
ACM Journal on Emerging Technologies in Computing Systems · 31.01.2023 · 19 (1) · 1-20
DOI: 10.1145/3555808 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Journal on Emerging Technologies in Computing Systems, online erschienen am 09.12.2022. Verfasst von Mojan Javaheripi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0). Gefördert von: NSF-CNS, NSF TILOS AI, Intel PrivateAI Collaborative Research Institute.
Abstract (Originalsprache)
We propose AccHashtag , the first framework for high-accuracy detection of fault-injection attacks on Deep Neural Networks (DNNs) with provable bounds on detection performance. Recent literature in fault-injection attacks shows the severe DNN accuracy degradation caused by bit flips. In this scenario, the attacker changes a few DNN weight bits during execution by injecting faults to the dynamic random-access memory (DRAM). To detect bit flips, AccHashtag extracts a unique signature from the benign DNN prior to deployment. The signature is used to validate the model’s integrity and verify the inference output on the fly. We propose a novel sensitivity analysis that identifies the most vulnerable DNN layers to the fault-injection attack. The DNN signature is constructed by encoding the weights in vulnerable layers using a low-collision hash function. During DNN inference, new hashes are extracted from the target layers and compared against the ground-truth signatures. AccHashtag incorporates a lightweight methodology that allows for real-time fault detection on embedded platforms. We devise a specialized compute core for AccHashtag on field-programmable gate arrays (FPGAs) to facilitate online hash generation in parallel to DNN execution. Extensive evaluations with the state-of-the-art bit-flip attack on various DNNs demonstrate the competitive advantage of AccHashtag in terms of both attack detection and execution overhead.
Förderung
- NSF-CNS
- NSF TILOS AI
- Intel PrivateAI Collaborative Research Institute
Zitieren
APA
Javaheripi, M., Chang, J., & Koushanfar, F. (2023). AccHashtag : Accelerated Hashing for Detecting Fault-Injection Attacks on Embedded Neural Networks. ACM Journal on Emerging Technologies in Computing Systems, 19(1), 1-20. https://doi.org/10.1145/3555808
Vancouver
Javaheripi M, Chang J, Koushanfar F. AccHashtag : Accelerated Hashing for Detecting Fault-Injection Attacks on Embedded Neural Networks. ACM Journal on Emerging Technologies in Computing Systems. 2023;19(1):1-20. doi:10.1145/3555808
BibTeX
@article{Javaheripi2023,
author = {Javaheripi, Mojan and Chang, Jung-Woo and Koushanfar, Farinaz},
title = {AccHashtag : Accelerated Hashing for Detecting Fault-Injection Attacks on Embedded Neural Networks},
journal = {ACM Journal on Emerging Technologies in Computing Systems},
year = {2023},
volume = {19},
number = {1},
pages = {1--20},
doi = {10.1145/3555808}
}
EUVIMED