Neural Network Laundering: Removing Black-Box Backdoor Watermarks from Deep Neural Networks.
William AikenHyoungshick KimSimon S. WooPublished in: CoRR (2020)
Keyphrases
- black box
- neural network
- black boxes
- digital images
- white box
- rule extraction
- hybrid systems
- back propagation
- pattern recognition
- integration testing
- trained neural networks
- artificial neural networks
- fault diagnosis
- fuzzy logic
- watermarking scheme
- activation function
- neural nets
- auto associative
- network architecture
- feed forward neural networks
- white box testing
- copyright protection
- neural network is trained
- trained neural network
- training algorithm
- feed forward
- neural network model
- hidden layer
- test cases
- genetic algorithm
- data sets
- training process
- recurrent neural networks
- radial basis function
- database