Conférence
Notice
Langue :
Anglais
Crédits
CNRS - Centre National de la Recherche Scientifique (Publication), INRIA (Institut national de recherche en informatique et automatique) (Publication), UNS (Publication), Marta Kwiatkowska (Intervention)
Conditions d'utilisation
Droit commun de la propriété intellectuelle
DOI : 10.60527/nhf8-pk76
Citer cette ressource :
Marta Kwiatkowska. Inria. (2017, 19 octobre). Safety Verification of Deep Neural Networks. [Vidéo]. Canal-U. https://doi.org/10.60527/nhf8-pk76. (Consultée le 25 avril 2025)
logo Inria
Inria - Institut national de recherche en sciences et technologies du numérique
S'abonner
Contacter

Safety Verification of Deep Neural Networks

Réalisation : 19 octobre 2017 - Mise en ligne : 27 octobre 2017
  • document 1 document 2 document 3
  • niveau 1 niveau 2 niveau 3
Descriptif

Deep neural networks have achieved impressive experimental results in image classification, but can surprisingly be unstable with respect to adversarial perturbations, that is, minimal changes to the input image that cause the network to misclassify it. With potential applications including perception modules and end-to-end controllers for self-driving cars, this raises concerns about their safety. This lecture will describe progress with developing a novel automated verification framework for deep neural networks to ensure safety of their classification decisions with respect to image manipulations, for example scratches or changes to camera angle or lighting conditions, that should not affect the classification. The techniques work directly with the network code and, in contrast to existing methods, can offer guarantees that adversarial examples are found if they exist. We implement the techniques using Z3 and evaluate them on state-of-the-art networks, including regularised and deep learning networks. We also compare against existing techniques to search for adversarial examples.

Intervention
Thème
Disciplines :

Sur le même thème