Neural network laundering: Removing black-box backdoor watermarks from deep neural networks.
William AikenHyoungshick KimSimon S. WooJungwoo RyooPublished in: Comput. Secur. (2021)
Keyphrases
- black box
- neural network
- digital images
- black boxes
- white box
- back propagation
- artificial neural networks
- rule extraction
- integration testing
- watermarking scheme
- pattern recognition
- fault diagnosis
- copyright protection
- neural network model
- hybrid systems
- white box testing
- trained neural networks
- recurrent neural networks
- test cases
- fuzzy logic
- multilayer perceptron
- activation function
- neural network is trained
- multi layer
- auto associative
- trained neural network
- hybrid neural network
- neural nets
- feedforward neural networks
- self organizing maps
- case study
- decision trees
- learning algorithm
- genetic algorithm