Randomization matters how to defend against strong adversarial attacks
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 …
attacks?} This paper tackles this question by adopting a game-theoretic point of view. We …
Mixed nash equilibria in the adversarial examples game
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 …
We study the open question of the existence of mixed Nash equilibria in the zero-sum game …
Nash equilibria and pitfalls of adversarial training in adversarial robustness games
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 …
paper, we study adversarial training as an alternating best-response strategy in a 2-player …
Adversarial attacks for mixtures of classifiers
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 …
improve robustness against adversarial attacks. However, it has been shown that existing …
Towards optimal randomized strategies in adversarial example game
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 …
challenge in many artificial intelligence applications. A recent line of work shows that the use …
On the role of randomization in adversarially robust classification
L Gnecco Heredia, MS Pydi… - Advances in …, 2023 - proceedings.neurips.cc
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 …
data. To defend against adversarial attacks, probabilistic classifiers have been proposed as …
Robustness verification for classifier ensembles
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 …
against arbitrary randomized attacks. Such attacks consist of a set of deterministic attacks …
On the role of randomization in adversarially robust classification
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 …
data. To defend against adversarial attacks, probabilistic classifiers have been proposed as …
Advances in Optimal Transport: Low-Rank Structures and Applications in Machine Learning
M Scetbon - 2023 - theses.hal.science
Recent advances in hardware, such as the development of highly-parallel accelerators, and
the growing permeabilitybetween computer science, statistics, optimization and applied …
the growing permeabilitybetween computer science, statistics, optimization and applied …
Scalable optimal classifiers for adversarial settings under uncertainty
B Roussillon, P Loiseau - International Conference on Decision and Game …, 2021 - Springer
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 …
class-1 data is generated by an attacker whose objective is not known to the defender—an …