Differentially private cross-silo federated learning

Mikko A. Heikkilä, Antti Koskela, Kana Shimizu +2

Strict privacy is of paramount importance in distributed machine learning. Federated learning, with the main idea of communicating only what is needed for learning, has been recent…