Stabilizing Inputs to Approximated Nonlinear Functions for Inference with Homomorphic Encryption in Deep Neural Networks.
Moustafa AboulAttaMatthias OssadnikSeyed-Ahmad AhmadiPublished in: CoRR (2019)
Keyphrases
- nonlinear functions
- neural network
- nonlinear systems
- hidden units
- rbf network
- homomorphic encryption
- input output
- fuzzy systems
- activation function
- privacy preserving
- data sharing
- back propagation
- radial basis function
- pattern recognition
- neural network model
- artificial neural networks
- fuzzy control
- basis functions
- fuzzy controller
- encryption scheme
- fuzzy logic
- control law
- feed forward neural networks
- fault diagnosis
- genetic algorithm
- neural nets
- training algorithm
- learning rate
- rbf neural network
- multilayer perceptron
- fuzzy sets
- nearest neighbor
- machine learning