DI-AA: An Interpretable White-box Attack for Fooling Deep Neural Networks.
Yixiang WangJiqiang LiuXiaolin ChangJianhua WangRicardo J. RodríguezPublished in: CoRR (2021)
Keyphrases
- white box
- neural network
- black box
- source code
- rule extraction
- fuzzy logic
- pattern recognition
- test data
- genetic algorithm
- artificial neural networks
- attack detection
- multi layer
- multilayer perceptron
- source code metrics
- deep learning
- neural nets
- radial basis function
- back propagation
- feed forward
- classification rules
- fault diagnosis
- fuzzy systems
- countermeasures
- secret key
- software development
- dos attacks
- feature extraction
- machine learning
- database