Privacy-Preserving Federated Learning

Co-authored UN paper exploring federated aggregation strategies with differential privacy and homomorphic encryption for national statistical offices.

Federated Learning · Differential Privacy · Python · Research

The challenge

National statistical offices hold sensitive data that cannot leave their premises, yet cross-border collaboration on machine learning models is needed to generate better statistics for the public good.

Approach

Key takeaway

Conducted under the United Nations PETLab. FedAvg and FedYogi are strong baselines that work well even without hyperparameter tuning. Differential privacy degrades weaker strategies less predictably — noise can accidentally help poorly performing models. Published at the UNECE Conference of European Statisticians, September 2023.