Towards Multi-timescale Online Monitoring of AI Models: Principles and Preliminary Results - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Towards Multi-timescale Online Monitoring of AI Models: Principles and Preliminary Results

Paul-Marie Raffi
  • Fonction : Auteur

Résumé

Online monitoring is an architectural pattern well-known to safety engineers, but it had to be adapted to AI technologies. In this paper, an innovative multi-time scale online monitoring architecture is presented. The main idea is to combine several monitoring timescales - Present- Time Monitoring (PTM), Near-Past Monitoring (NPM), and Near-Future Monitoring (NFM) - on different monitoring assets (inputs, internal states, and outputs of the AI model) to ensure a high anomaly detection rate by design of the online monitor.
Fichier principal
Vignette du fichier
44.pdf (2.14 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY - Paternité

Dates et versions

hal-04240929 , version 1 (13-10-2023)

Identifiants

  • HAL Id : hal-04240929 , version 1

Citer

Fateh Kaakai, Paul-Marie Raffi. Towards Multi-timescale Online Monitoring of AI Models: Principles and Preliminary Results. SafeAI, AAAI’s Workshop on Artificial Intelligence Safety, Feb 2023, Washinghton, DC, United States. ⟨hal-04240929⟩
37 Consultations
28 Téléchargements

Partager

Gmail Facebook X LinkedIn More