English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The wide deployment of Deep Neural Networks (DNN) in high-performance cloudcomputing platforms brought to light multi-tenant cloud field-programmable gatearrays (FPGA) as a popular choice of accelerator to boost performance due toits hardware reprogramming flexibility. Such a multi-tenant FPGA setup for DNNacceleration potentially exposes DNN interference tasks under severe threatfrom malicious users. This work, to the best of our knowledge, is the first toexplore DNN model vulnerabilities in multi-tenant FPGAs. We propose a noveladversarial attack framework: Deep-Dup, in which the adversarial tenant caninject adversarial faults to the DNN model in the victim tenant of FPGA.Specifically, she can aggressively overload the shared power distributionsystem of FPGA with malicious power-plundering circuits, achieving adversarialweight duplication (AWD) hardware attack that duplicates certain DNN weightpackages during data transmission between off-chip memory and on-chip buffer,to hijack the DNN function of the victim tenant. Further, to identify the mostvulnerable DNN weight packages for a given malicious objective, we propose ageneric vulnerable weight package searching algorithm, called ProgressiveDifferential Evolution Search (P-DES), which is, for the first time, adaptiveto both deep learning white-box and black-box attack models. The proposedDeep-Dup is experimentally validated in a developed multi-tenant FPGAprototype, for two popular deep learning applications, i.e., Object Detectionand Image Classification. Successful attacks are demonstrated in six popularDNN architectures (e.g., YOLOv2, ResNet-50, MobileNet, etc.)

Deep-Dup: An Adversarial Weight Duplication Attack Framework to Crush Deep Neural Network in Multi-Tenant FPGA