A CNN trained on AudioMNIST is shown to be vulnerable to FGSM, PGD, CW, and data poisoning attacks, while two off-the-shelf defense libraries fail to restore accuracy.
Goodfellow, Rob Fergus // 2nd Intern ational Conference on Learning Representa- tions, ICLR 2014
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Attacks on the neural network and defense methods
A CNN trained on AudioMNIST is shown to be vulnerable to FGSM, PGD, CW, and data poisoning attacks, while two off-the-shelf defense libraries fail to restore accuracy.