DNN-Defender: An in-DRAM Deep Neural Network Defense Mechanism for Adversarial Weight Attack. (arXiv:2305.08034v1 [cs.CR])

With deep learning deployed in many security-sensitive areas, machine
learning security is becoming progressively important. Recent studies
demonstrate attackers can exploit system-level techniques exploiting the
RowHammer vulnerability of DRAM to deterministically and precisely flip bits in
Deep Neural Networks (DNN) model weights to affect inference accuracy. The
existing defense mechanisms are software-based, such as weight reconstruction
requiring expensive training overhead or performance degradation. On the other
hand, generic hardware-based victim-/aggressor-focused mechanisms impose
expensive hardware overheads and preserve the spatial connection …

accuracy adversarial attack attackers bits deep learning defender defense dram exploit exploiting important machine machine learning network network defense networks neural network neural networks precisely rowhammer security studies system techniques vulnerability