Federated Learning with Bayesian Differential Privacy
This work addresses privacy concerns in federated learning for applications like medical image classification, offering incremental improvements over existing differential privacy methods.
The paper tackled the problem of providing formal privacy guarantees in federated learning by using Bayesian differential privacy, resulting in sharper privacy loss bounds with a privacy budget below 1 at the client level and below 0.1 at the instance level, while improving model accuracy and reducing communication rounds.
We consider the problem of reinforcing federated learning with formal privacy guarantees. We propose to employ Bayesian differential privacy, a relaxation of differential privacy for similarly distributed data, to provide sharper privacy loss bounds. We adapt the Bayesian privacy accounting method to the federated setting and suggest multiple improvements for more efficient privacy budgeting at different levels. Our experiments show significant advantage over the state-of-the-art differential privacy bounds for federated learning on image classification tasks, including a medical application, bringing the privacy budget below 1 at the client level, and below 0.1 at the instance level. Lower amounts of noise also benefit the model accuracy and reduce the number of communication rounds.