Do Gradient Inversion Attacks Make Federated Learning Unsafe?

Publication
arXiv preprint arXiv:2202.06924