Private Federated Learning with Autotuned Compression
Enayat Ullah,u00a0Christopher A. Choquette-Choo,u00a0Peter Kairouz,u00a0Sewoong Oh
We propose new techniques for reducing communication in private federated learning without the need for setting or tuning compression rates. Our on-the-fly methods automatically adjust the compression rate based on the error induced during training, while maintaining provable privacy guarantees through the use of secure aggregation and differential privacy. Our techniques are provably instance-optimal for mean estimation, meaning that they can adapt to the u201chardness of the problemu201d with minimal interactivity. We demonstrate the effectiveness of our approach on real-world datasets by achieving favorable compression rates without the need for tuning.
