Timezone: »

The Distributed Discrete Gaussian Mechanism for Federated Learning with Secure Aggregation
Peter Kairouz · Ziyu Liu · Thomas Steinke

Thu Jul 22 09:00 PM -- 11:00 PM (PDT) @

We consider training models on private data that are distributed across user devices. To ensure privacy, we add on-device noise and use secure aggregation so that only the noisy sum is revealed to the server. We present a comprehensive end-to-end system, which appropriately discretizes the data and adds discrete Gaussian noise before performing secure aggregation. We provide a novel privacy analysis for sums of discrete Gaussians and carefully analyze the effects of data quantization and modular summation arithmetic. Our theoretical guarantees highlight the complex tension between communication, privacy, and accuracy. Our extensive experimental results demonstrate that our solution is essentially able to match the accuracy to central differential privacy with less than 16 bits of precision per value.

Author Information

Peter Kairouz (Google)
Ziyu Liu (Google Research)
Thomas Steinke (Google)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors