NECO: NEURAL COLLAPSE BASED OUT-OF-DISTRIBUTION DETECTION - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

NECO: NEURAL COLLAPSE BASED OUT-OF-DISTRIBUTION DETECTION

Résumé

Detecting out-of-distribution (OOD) data is a critical challenge in machine learning due to model overconfidence, often without awareness of their epistemological limits. We hypothesize that "neural collapse", a phenomenon affecting in-distribution data for models trained beyond loss convergence, also influences OOD data. To benefit from this interplay, we introduce NECO, a novel post-hoc method for OOD detection, which leverages the geometric properties of "neural collapse" and of principal component spaces to identify OOD data. Our extensive experiments demonstrate that NECO achieves state-of-the-art results on both small and large-scale OOD detection tasks while exhibiting strong generalization capabilities across different network architectures. Furthermore, we provide a theoretical explanation for the effectiveness of our method in OOD detection. Code is available at https://gitlab.com/drti/neco.
Fichier principal
Vignette du fichier
NECO_ICLR_final_.pdf (6.43 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04480548 , version 1 (05-03-2024)

Identifiants

  • HAL Id : hal-04480548 , version 1

Citer

Mouin Ben Ammar, Nacim Belkhir, Sebastien Popescu, Antoine Manzanera, Gianni Franchi. NECO: NEURAL COLLAPSE BASED OUT-OF-DISTRIBUTION DETECTION. The Twelfth International Conference on Learning Representations, May 2024, Vienne (AUT), France. ⟨hal-04480548⟩
6 Consultations
5 Téléchargements

Partager

Gmail Facebook X LinkedIn More