# gradient-leakage

Federated Learning Attacks

Attacking federated learning through model update poisoning, gradient leakage, free-rider attacks, and Byzantine fault exploitation.