Skip to yearly menu bar Skip to main content


Private Federated Learning with Autotuned Compression

Enayat Ullah · Christopher Choquette-Choo · Peter Kairouz · Sewoong Oh

Exhibit Hall 1 #406


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 ``hardness of the problem'' with minimal interactivity. We demonstrate the effectiveness of our approach on real-world datasets by achieving favorable compression rates without the need for tuning.

Chat is not available.