Existence of Adversarial Examples for Random Convolutional Networks via Isoperimetric Inequalities on $\mathbb{SO}(d)$

Abstract

We show that adversarial examples exist for various random convolutional networks, and furthermore, that this is a relatively simple consequence of the isoperimetric inequality on the special orthogonal group $\mathbb{SO}(d)$. This extends and simplifies a recent line of work which shows similar results for random fully connected networks.

Cite

Text

Daniely. "Existence of Adversarial Examples for Random Convolutional Networks via  Isoperimetric Inequalities on $\mathbb{SO}(d)$." Proceedings of Thirty Eighth Conference on Learning Theory, 2025.

Markdown

[Daniely. "Existence of Adversarial Examples for Random Convolutional Networks via  Isoperimetric Inequalities on $\mathbb{SO}(d)$." Proceedings of Thirty Eighth Conference on Learning Theory, 2025.](https://mlanthology.org/colt/2025/daniely2025colt-existence/)

BibTeX

@inproceedings{daniely2025colt-existence,
  title     = {{Existence of Adversarial Examples for Random Convolutional Networks via  Isoperimetric Inequalities on $\mathbb{SO}(d)$}},
  author    = {Daniely, Amit},
  booktitle = {Proceedings of Thirty Eighth Conference on Learning Theory},
  year      = {2025},
  pages     = {1368-1379},
  volume    = {291},
  url       = {https://mlanthology.org/colt/2025/daniely2025colt-existence/}
}