On Neural Networks Redundancy and Diversity for Their Use in Safety-Critical Systems

An increasing number of critical functionalities integrated in embedded critical systems rely on deep learning (DL) technology. This article summarizes certain key aspects of DL’s intrinsic stochastic and training-data-dependent nature that are at odds with current domain-specific functional safety...

Descripción completa

Detalles Bibliográficos
Autores: Brando, Axel, Serra, Isabel, Mezzetti, Enrico|||0000-0002-1886-2931, Cazorla Almeida, Francisco Javier, Perez Cerrolaza, Jon, Abella Ferrer, Jaume|||0000-0001-7951-4028
Tipo de recurso: artículo
Fecha de publicación:2023
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/387765
Acceso en línea:https://hdl.handle.net/2117/387765
https://dx.doi.org/10.1109/MC.2023.3236523
Access Level:acceso abierto
Palabra clave:Deep learning (Machine learning)
Neural networks (Computer science)
Deep learning
Redundancy
Neural networks
Safety
Embedded systems
Contingency management
Intel·ligència artificial
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
Descripción
Sumario:An increasing number of critical functionalities integrated in embedded critical systems rely on deep learning (DL) technology. This article summarizes certain key aspects of DL’s intrinsic stochastic and training-data-dependent nature that are at odds with current domain-specific functional safety standards. We exemplify how redundancy and diversity of neural networks can help to reconcile DL technology and functional safety requirements.