Impact of Low-Bitwidth Quantization on the Adversarial Robustness for Embedded Neural Networks.
Rémi BernhardPierre-Alain MoëllicJean-Max DutertrePublished in: CW (2019)
Keyphrases
- neural network
- digital image watermarking scheme
- embedded systems
- pattern recognition
- artificial neural networks
- genetic algorithm
- quantization error
- back propagation
- fuzzy logic
- neural network model
- recurrent neural networks
- self organizing maps
- network architecture
- data mining
- computational efficiency
- control system
- high levels
- multi agent
- high robustness