Efficient Statistical Assessment of Neural Network Corruption Robustness - Processus stochastiques Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Efficient Statistical Assessment of Neural Network Corruption Robustness

Résumé

We quantify the robustness of a trained network to input uncertainties with a stochastic simulation inspired by the field of Statistical Reliability Engineering. The robustness assessment is cast as a statistical hypothesis test: the network is deemed as locally robust if the estimated probability of failure is lower than a critical level. The procedure is based on an Importance Splitting simulation generating samples of rare events. We derive theoretical guarantees that are nonasymptotic w.r.t. sample size. Experiments tackling large scale networks outline the efficiency of our method making a low number of calls to the network function.
Fichier principal
Vignette du fichier
Robustness_Estimation_with_Last_Particle__NeurIPS_pre_print_-6.pdf (343.26 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03407011 , version 1 (28-10-2021)
hal-03407011 , version 2 (29-10-2021)

Identifiants

  • HAL Id : hal-03407011 , version 2

Citer

Karim Tit, Teddy Furon, Mathias Rousset. Efficient Statistical Assessment of Neural Network Corruption Robustness. NeurIPS 2021 - 35th Conference on Neural Information Processing Systems, Dec 2021, Sydney (virtual), Australia. ⟨hal-03407011v2⟩
236 Consultations
236 Téléchargements

Partager

Gmail Facebook X LinkedIn More