Abstract\emph {Is there a classifier that ensures optimal robustness against all adversarial
attacks?} This paper tackles this question by adopting a game-theoretic point of view. We …

This paper tackles the problem of adversarial examples from a game theoretic point of view.
We study the open question of the existence of mixed Nash equilibria in the zero-sum game …

Adversarial training is a standard technique for training adversarially robust models. In this
paper, we study adversarial training as an alternating best-response strategy in a 2-player …

Mixtures of classifiers (aka randomized ensembles) have been proposed as a way to
improve robustness against adversarial attacks. However, it has been shown that existing …

The vulnerability of deep neural network models to adversarial example attacks is a practical
challenge in many artificial intelligence applications. A recent line of work shows that the use …

Deep neural networks are known to be vulnerable to small adversarial perturbations in test
data. To defend against adversarial attacks, probabilistic classifiers have been proposed as …

We give a formal verification procedure that decides whether a classifier ensemble is robust
against arbitrary randomized attacks. Such attacks consist of a set of deterministic attacks …

Deep neural networks are known to be vulnerable to small adversarial perturbations in test
data. To defend against adversarial attacks, probabilistic classifiers have been proposed as …

Recent advances in hardware, such as the development of highly-parallel accelerators, and
the growing permeabilitybetween computer science, statistics, optimization and applied …

We consider the problem of finding optimal classifiers in an adversarial setting where the
class-1 data is generated by an attacker whose objective is not known to the defender—an …