Viet-Anh on Software Logo

What is: Fast Minimum-Norm Attack?

SourceFast Minimum-norm Adversarial Attacks through Adaptive Norm Constraints
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

Fast Minimum-Norm Attack, or FNM, is a type of adversarial attack that works with different p\ell_{p}-norm perturbation models (p=0,1,2,p=0,1,2,\infty), is robust to hyperparameter choices, does not require adversarial starting points, and converges within few lightweight steps. It works by iteratively finding the sample misclassified with maximum confidence within an p\ell_{p}-norm constraint of size ϵ\epsilon, while adapting ϵ\epsilon to minimize the distance of the current sample to the decision boundary.