Federated Learning with Local Differential Privacy: Trade-offs between Privacy, Utility, and Communication
This work addresses the problem of balancing privacy, utility, and communication efficiency in federated learning for practitioners and researchers concerned with data privacy.
This paper explores the trade-offs between privacy, utility, and communication in federated learning (FL) with local differential privacy (LDP) using Gaussian mechanisms. It defines query sensitivity as a variable and applies a tighter privacy accounting method, demonstrating that for a given privacy level, their approach yields significantly larger utility and a smaller transmission rate compared to existing methods.
Federated learning (FL) allows to train a massive amount of data privately due to its decentralized structure. Stochastic gradient descent (SGD) is commonly used for FL due to its good empirical performance, but sensitive user information can still be inferred from weight updates shared during FL iterations. We consider Gaussian mechanisms to preserve local differential privacy (LDP) of user data in the FL model with SGD. The trade-offs between user privacy, global utility, and transmission rate are proved by defining appropriate metrics for FL with LDP. Compared to existing results, the query sensitivity used in LDP is defined as a variable and a tighter privacy accounting method is applied. The proposed utility bound allows heterogeneous parameters over all users. Our bounds characterize how much utility decreases and transmission rate increases if a stronger privacy regime is targeted. Furthermore, given a target privacy level, our results guarantee a significantly larger utility and a smaller transmission rate as compared to existing privacy accounting methods.