Efficient Evaluation of Activation Functions over Encrypted Data
Patricia Thaine, Sergey V Gorbunov, Gerald M. Penn · 2019
We describe a method for approximating any bounded activation function given encrypted input data. The utility of our method is exemplified by simulating it within two typical machine learning tasks: namely, a Variational Autoencoder that learns a latent representation of MNIST data, and an MNIST image classifier.