#développement sûr et embarquabilité
#Fair learning #Robotique et IA
The work proposed consists in identifying lack of certification standards for new phenomena and proposing new approaches to help develop and certify AI applications, e.g., integrating the notion of algorithm failure, non deterministic and unpredictable behaviour, run time services to detect faulty behavior in sophisticated AI systems.
IA acceptable, certifiable & collaborative
Porteur :
Claire Pagetti, Onera
Contact
Claire.Pagetti@onera.fr
Equipe: Kevin Delmas
Charles Lesire-Cabaniols
Thomas Carles
Jérémie Guiochet
Aurélien Plyer