Fault Tolerant Deep Neural Networks for Detection of Unrecognizable Situations - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Fault Tolerant Deep Neural Networks for Detection of Unrecognizable Situations

Résumé

Deep Neural Networks are achieving great success in various fields. However, their use remains limited to non critical applications because their behavior is unpredictable and unsafe. In this paper we propose some fault tolerant approaches based on diversifying learning in order to improve DNNs dependability and particularly safety. Our main goal is to increase trust in the outcome of deep learning mechanisms by recognizing the unlearned inputs and preventing misclassification.
Fichier principal
Vignette du fichier
safeProcess18-rhazali.pdf (617.42 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01996374 , version 1 (28-01-2019)

Identifiants

Citer

Kaoutar Rhazali, Benjamin Lussier, Walter Schön, Stéphane Géronimi. Fault Tolerant Deep Neural Networks for Detection of Unrecognizable Situations. 10th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS 2018), Aug 2018, Warsaw, Poland. pp.31-37, ⟨10.1016/j.ifacol.2018.09.525⟩. ⟨hal-01996374⟩
84 Consultations
88 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More