Out-of-Distribution Detection Using Deep Neural Network Latent Space Uncertainty - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Out-of-Distribution Detection Using Deep Neural Network Latent Space Uncertainty

Fabio Arnez
Ansgar Radermacher
  • Fonction : Auteur
  • PersonId : 996139
François Terrier
  • Fonction : Auteur
  • PersonId : 1133688

Résumé

As automated systems increasingly incorporate deep neural networks (DNNs) to perform safety-critical tasks, confidence representation and uncertainty estimation in DNN predictions have become useful and essential to represent DNN ignorance. Predictive uncertainty has often been used to identify samples that can lead to wrong predictions with high confidence, i.e., Out-of-Distribution (OoD) detection. However, predictive uncertainty estimation at the output of a DNN might fail for OoD detection in computer vision tasks such as semantic segmentation due to the lack of information about semantic structures and contexts. We propose using the DNN uncertainty from intermediate latent representations to overcome this problem. Our experiments show promising results in OoD detection for the semantic segmentation task.
Fichier non déposé

Dates et versions

hal-04486388 , version 1 (01-03-2024)

Identifiants

  • HAL Id : hal-04486388 , version 1

Citer

Fabio Arnez, Ansgar Radermacher, François Terrier. Out-of-Distribution Detection Using Deep Neural Network Latent Space Uncertainty. 37th AAAI Conference on Artificial Intelligence: SafeAI 2023 workshop, Feb 2023, Washigton, D.C., United States. ⟨hal-04486388⟩
14 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More