Privacy-preserving gradient-based fair federated learning

Janis Adamek, Moritz Schulze Darup

Federated learning (FL) schemes allow multiple participants to collaboratively train neural networks without the need to directly share the underlying data.However, in early scheme…