Towards Consistency in Adversarial Classification - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Towards Consistency in Adversarial Classification

Vers la consistence en classification adversariale

Résumé

In this paper, we study the problem of consistency in the context of adversarial examples. Specifically, we tackle the following question: Can surrogate losses still be used as a proxy for minimizing the 0/1 loss in the presence of an adversary that alters the inputs at test-time? Different from the standard classification task, this question cannot be reduced to a point-wise minimization problem, and calibration needs not to be sufficient to ensure consistency. In this paper, we expose some pathological behaviors specific to the adversarial problem, and show that no convex surrogate loss can be consistent or calibrated in this context. It is therefore necessary to design another class of surrogate functions that can be used to solve the adversarial consistency issue. As a first step towards designing such a class, we identify sufficient and necessary conditions for a surrogate loss to be calibrated in both the adversarial and standard settings. Finally, we give some directions for building a class of losses that could be consistent in the adversarial framework.
Fichier principal
Vignette du fichier
8998_towards_consistency_in_adversa.pdf (614.92 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04014745 , version 1 (04-03-2023)

Identifiants

  • HAL Id : hal-04014745 , version 1

Citer

Laurent Meunier, Raphael Ettedgui, Rafaël Pinot, Yann Chevaleyre, Jamal Atif. Towards Consistency in Adversarial Classification. 36th Conference on Neural Information Processing Systems (NeurIPS 2022), Nov 2022, Virtual conference, United States. ⟨hal-04014745⟩
45 Consultations
8 Téléchargements

Partager

Gmail Facebook X LinkedIn More