Robustness of Bayesian Neural Networks to White-Box Adversarial Attacks.
Adaku UchenduDaniel CampoyChristopher MenartAlexandra HildenbrandtPublished in: CoRR (2021)
Keyphrases
- neural network
- white box
- black box
- digital image watermarking
- artificial neural networks
- source code
- semi fragile watermarking
- bayesian networks
- recurrent neural networks
- geometric attacks
- bayesian inference
- pattern recognition
- maximum likelihood
- image watermarking scheme
- countermeasures
- neural network model
- back propagation
- fuzzy logic
- multi agent
- genetic algorithm
- source code metrics
- probabilistic model
- image watermarking
- watermark embedding
- malicious users
- watermarking scheme
- multilayer perceptron
- posterior probability
- test data
- error rate