English

Defensive Distillation is Not Robust to Adversarial Examples

Cryptography and Security 2016-07-18 v1 Computer Vision and Pattern Recognition

Abstract

We show that defensive distillation is not secure: it is no more resistant to targeted misclassification attacks than unprotected neural networks.

Cite

@article{arxiv.1607.04311,
  title  = {Defensive Distillation is Not Robust to Adversarial Examples},
  author = {Nicholas Carlini and David Wagner},
  journal= {arXiv preprint arXiv:1607.04311},
  year   = {2016}
}
R2 v1 2026-06-22T14:55:15.468Z