Adversarial Consistency and the Uniqueness of the Adversarial Bayes Classifier

arXiv:2404.17358v1 Announce Type: new
Abstract: Adversarial training is a common technique for learning robust classifiers. Prior work showed that convex surrogate losses are not statistically consistent in the adversarial context -- or in other words, a minimizing sequence of the adversarial surrogate risk will not necessarily minimize the adversarial classification error. We connect the consistency of adversarial surrogate losses to properties of minimizers to the adversarial classification risk, known as \emph{adversarial Bayes classifiers}. Specifically, under reasonable distributional assumptions, a convex …

abstract adversarial adversarial training arxiv bayes classification classifier classifiers consistent context cs.lg losses math.st prior risk robust stat.ml stat.th training type will words work