ISimDL: Importance Sampling-Driven Acceleration of Fault Injection Simulations for Evaluating the Robustness of Deep Learning.
Alessio ColucciAndreas SteiningerMuhammad ShafiquePublished in: CoRR (2023)
Keyphrases
- deep learning
- importance sampling
- fault injection
- monte carlo
- unsupervised learning
- markov chain
- kalman filter
- java card
- particle filter
- machine learning
- approximate inference
- particle filtering
- fault model
- markov chain monte carlo
- weakly supervised
- mental models
- posterior distribution
- supervised learning
- object recognition
- latent variables
- pairwise