MDSE attack uses dynamic multi-surrogate gradient estimation to create adversarial examples that simultaneously fool SNNs, ViTs, and CNNs, with reported gains up to 91.4% on ensembles and 3x on adversarially trained SNNs versus Auto-PGD.
H.; Dimou, G.; Joshi, P.; Imam, N.; Jain, S.; et al
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2022 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Attacking the Spike: On the Transferability and Security of Spiking Neural Networks to Adversarial Examples
MDSE attack uses dynamic multi-surrogate gradient estimation to create adversarial examples that simultaneously fool SNNs, ViTs, and CNNs, with reported gains up to 91.4% on ensembles and 3x on adversarially trained SNNs versus Auto-PGD.