Characterizing a Neutron-Induced Fault Model for Deep Neural Networks.
Fernando Fernandes dos SantosAngeliki KritikakouJosie Esteban Rodriguez CondiaJuan-David Guerrero-BalagueraMatteo Sonza ReordaOlivier SentieysPaolo RechPublished in: CoRR (2022)
Keyphrases
- fault model
- neural network
- safety analysis
- pattern recognition
- fault injection
- artificial neural networks
- multi layer
- back propagation
- recurrent neural networks
- fuzzy logic
- multilayer perceptron
- genetic algorithm
- neural nets
- feed forward
- image quality
- training process
- model based diagnosis
- constraint programming
- np hard
- search algorithm
- reinforcement learning