First Provable Guarantees for Practical Private FL: Beyond Restrictive Assumptions
This work addresses the challenge of making private federated learning more practical for real-world applications by moving beyond restrictive assumptions, though it is incremental in building on existing methods.
The paper tackled the problem of providing practical differentially private federated learning by introducing Fed-$α$-NormEC, a framework that offers provable convergence and privacy guarantees under standard assumptions while supporting features like multiple local updates and partial client participation, with experimental validation on deep learning tasks.
Federated Learning (FL) enables collaborative training on decentralized data. Differential privacy (DP) is crucial for FL, but current private methods often rely on unrealistic assumptions (e.g., bounded gradients or heterogeneity), hindering practical application. Existing works that relax these assumptions typically neglect practical FL features, including multiple local updates and partial client participation. We introduce Fed-$α$-NormEC, the first differentially private FL framework providing provable convergence and DP guarantees under standard assumptions while fully supporting these practical features. Fed-$α$-NormE integrates local updates (full and incremental gradient steps), separate server and client stepsizes, and, crucially, partial client participation, which is essential for real-world deployment and vital for privacy amplification. Our theoretical guarantees are corroborated by experiments on private deep learning tasks.