A Survey on Methods for the Safety Assurance of Machine Learning Based Systems - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

A Survey on Methods for the Safety Assurance of Machine Learning Based Systems

Résumé

Methods for safety assurance suggested by the ISO 26262 automotive functional safety standard are not sufficient for applications based on machine learning (ML). We provide a structured, certification oriented overview on available methods supporting the safety argumen-tation of a ML based system. It is sorted into life-cycle phases, and maturity of the approach as well as applicability to different ML types are collected. From this we deduce current open challenges: powerful solvers, inclusion of expert knowledge, validation of data representativity and model diversity, and model introspection with provable guarantees.
Fichier principal
Vignette du fichier
ERTS2020_paper_11.pdf (285.53 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02442819 , version 1 (16-01-2020)

Identifiants

  • HAL Id : hal-02442819 , version 1

Citer

Gesina Schwalbe, Martin Schels. A Survey on Methods for the Safety Assurance of Machine Learning Based Systems. 10th European Congress on Embedded Real Time Software and Systems (ERTS 2020), Jan 2020, Toulouse, France. ⟨hal-02442819⟩

Collections

ERTS2020
1165 Consultations
2175 Téléchargements

Partager

Gmail Facebook X LinkedIn More