Blog - Data Science Current

Distributed differential privacy for federated learning

Google Research AI blog

MARCH 2, 2023

While offering substantial additional protections, a fully malicious server might still be able to get around the DDP guarantees either by manipulating the public key exchange of SecAgg or by injecting a sufficient number of "fake" malicious clients that don’t add the prescribed noise into the aggregation pool.

ML ML Machine Learning Machine Learning

Data Science Current

Distributed differential privacy for federated learning

Webinars

Stay Connected