Deeptheft: Stealing dnn model architectures through power side channel

Y Gao, H Qiu, Z Zhang, B Wang, H Ma… - … IEEE Symposium on …, 2024 - ieeexplore.ieee.org
Deep Neural Network (DNN) models are often deployed in resource-sharing clouds as
Machine Learning as a Service (MLaaS) to provide inference services. To steal model …

Learning privacy-preserving graph convolutional network with partially observed sensitive attributes

H Hu, L Cheng, JP Vap, M Borowczak - Proceedings of the ACM Web …, 2022 - dl.acm.org
Recent studies have shown Graph Neural Networks (GNNs) are extremely vulnerable to
attribute inference attacks. To tackle this challenge, existing privacy-preserving GNNs …

Reverse engineering neural network folding with remote FPGA power analysis

V Meyers, D Gnad, M Tahoori - 2022 IEEE 30th Annual …, 2022 - ieeexplore.ieee.org
Specialized hardware accelerators in the form of FPGAs are widely being used for neural
network implementations. By that, they also become the target of power analysis attacks that …

[PDF][PDF] SoK: neural network extraction through physical side channels

P Horváth, D Lauret, Z Liu, L Batina - … of the 33rd USENIX Conference on …, 2024 - usenix.org
SoK Neural Network Extraction-USENIX Presentation Page 1 SoK: Neural Network Extraction
Through Physical Side Channels 15.08.2024 Péter Horváth, Dirk Lauret, Zhuoran Liu, and …

SoK: Model Reverse Engineering Threats for Neural Network Hardware

S Potluri, F Koushanfar - Cryptology ePrint Archive, 2024 - eprint.iacr.org
There has been significant progress over the past seven years in model reverse engineering
(RE) for neural network (NN) hardware. Although there has been systematization of …

Dual-Rail Precharge Logic-Based Side-Channel Countermeasure for DNN Systolic Array

L Wu, L Wu, X Zhang, M Chinbat - IEEE Transactions on Very …, 2024 - ieeexplore.ieee.org
Deep neural network (DNN) accelerators are widely used in cloud-edge-end and other
application scenarios. Researchers recently focused on extracting secret information from …

Transition Recovery Attack on Embedded State Machines Using Power Analysis

C Carper, A Robins… - 2022 IEEE 40th …, 2022 - ieeexplore.ieee.org
Modern embedded systems are ever present within our daily lives. Such devices remain
vulnerable to Differential Power Analysis via side-channel attacks, which provide a powerful …

TP-NET: Training Privacy-Preserving Deep Neural Networks under Side-Channel Power Attacks

H Hu, J Gegax-Randazzo, C Carper… - … Symposium on Smart …, 2022 - ieeexplore.ieee.org
Privacy in deep learning is receiving tremendous attention with its wide applications in
industry and academics. Recent studies have shown the internal structure of a deep neural …

Power Analysis Side-Channel Attacks on Same and Cross-Device Settings: A Survey of Machine Learning Techniques

A Ghimire, VV Baligodugula, F Amsaad - IFIP International Internet of …, 2023 - Springer
Abstract Systems that use secret keys or personal details are seriously at risk from side-
channel attacks, especially if they rely on power analysis. Attackers can use unintentional …

Hardware Trojan Key-Corruption Detection with Automated Neural Architecture Search

F Mezzarapa, J Goodrich, A Robins… - IFIP International Internet …, 2024 - Springer
This work presents a model hardware trojan which intermittently is capable of corrupting an
encryption operation occurring on a device. It asks whether this trojan can be detected via …