SIT: Stochastic Input Transformation to Defend Against Adversarial Attacks on Deep Neural Networks.
Amira GuesmiIhsen AlouaniMouna BakloutiTarek FrikhaMohamed AbidPublished in: IEEE Des. Test (2022)
Keyphrases
- neural network
- pattern recognition
- desired output
- artificial neural networks
- input data
- back propagation
- self organizing maps
- activation function
- feed forward
- malicious attacks
- monte carlo
- multilayer perceptron
- recurrent neural networks
- hidden units
- stochastic optimization
- deep learning
- malicious users
- feed forward neural networks
- network architecture
- watermarking scheme
- neural nets
- neural network model
- fuzzy logic
- multi agent