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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Daniely, Amit
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909649250287616
author Daniely, Amit
author_facet Daniely, Amit
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12613
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Existence of Adversarial Examples for Random Convolutional Networks via Isoperimetric Inequalities on $\mathbb{so}(d)$
Daniely, Amit
Machine Learning
Probability
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.
title Existence of Adversarial Examples for Random Convolutional Networks via Isoperimetric Inequalities on $\mathbb{so}(d)$
topic Machine Learning
Probability
url https://arxiv.org/abs/2506.12613