Federated Cubic Regularized Newton Learning with Sparsification-amplified Differential Privacy
This addresses privacy and efficiency challenges in federated learning for distributed data applications, representing an incremental improvement through integration of existing techniques.
The paper tackles privacy leakage and communication bottlenecks in federated learning by introducing DP-FCRN, a federated cubic-regularized Newton method with sparsification-amplified differential privacy, achieving lower iteration complexity than first-order methods and reducing communication costs while maintaining privacy.
This paper investigates the use of the cubic-regularized Newton method within a federated learning framework while addressing two major concerns that commonly arise in federated learning: privacy leakage and communication bottleneck. We introduce a federated learning algorithm called Differentially Private Federated Cubic Regularized Newton (DP-FCRN). By leveraging second-order techniques, our algorithm achieves lower iteration complexity compared to first-order methods. We also incorporate noise perturbation during local computations to ensure privacy. Furthermore, we employ sparsification in uplink transmission, which not only reduces the communication costs but also amplifies the privacy guarantee. Specifically, this approach reduces the necessary noise intensity without compromising privacy protection. We analyze the convergence properties of our algorithm and establish the privacy guarantee. Finally, we validate the effectiveness of the proposed algorithm through experiments on a benchmark dataset.