Mixed Nash Equilibria in the Adversarial Examples Game - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Mixed Nash Equilibria in the Adversarial Examples Game

Laurent Meunier
  • Fonction : Auteur
  • PersonId : 1080147
Meyer Scetbon
  • Fonction : Auteur
Rafaël Pinot
  • Fonction : Auteur
  • PersonId : 1026883
Jamal Atif

Résumé

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 formed by the attacker and the classifier. While previous works usually allow only one player to use randomized strategies, we show the necessity of considering randomization for both the classifier and the attacker. We demonstrate that this game has no duality gap, meaning that it always admits approximate Nash equilibria. We also provide the first optimization algorithms to learn a mixture of a finite number of classifiers that approximately realizes the value of this game, i.e. procedures to build an optimally robust randomized classifier.
Fichier principal
Vignette du fichier
meunier21a.pdf (1.21 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03916826 , version 1 (31-12-2022)

Identifiants

  • HAL Id : hal-03916826 , version 1

Citer

Laurent Meunier, Meyer Scetbon, Rafaël Pinot, Jamal Atif, Yann Chevaleyre. Mixed Nash Equilibria in the Adversarial Examples Game. International Conference on Machine Learning (ICML), Aug 2021, paris, France. ⟨hal-03916826⟩
9 Consultations
22 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More