Cao, X.; Jia, J.; Gong, N. Z. Provably Secure Federated Learning Against Malicious Clients. AAAI 2021, 35, 6885-6893.