Effect of Ambient-Intrinsic Dimension Gap on Adversarial Vulnerability

arXiv:2403.03967v1 Announce Type: cross
Abstract: The existence of adversarial attacks on machine learning models imperceptible to a human is still quite a mystery from a theoretical perspective. In this work, we introduce two notions of adversarial attacks: natural or on-manifold attacks, which are perceptible by a human/oracle, and unnatural or off-manifold attacks, which are not. We argue that the existence of the off-manifold attacks is a natural consequence of the dimension gap between the intrinsic and ambient dimensions of the …

adversarial adversarial attacks ambient arxiv attacks cs.cr cs.lg effect gap human machine machine learning machine learning models manifold natural oracle perspective stat.ml vulnerability work