Privacy-preserving Federated Primal-dual Learning for Non-convex and Non-smooth Problems with Model Sparsification
This addresses communication efficiency and privacy protection challenges in federated learning for applications with complex loss functions, representing an incremental improvement over existing FL algorithms.
The paper tackled federated learning problems with non-convex and non-smooth loss functions by proposing a novel federated primal-dual algorithm with bidirectional model sparsification and differential privacy, achieving superior performance over state-of-the-art methods in experiments on real-world data.
Federated learning (FL) has been recognized as a rapidly growing research area, where the model is trained over massively distributed clients under the orchestration of a parameter server (PS) without sharing clients' data. This paper delves into a class of federated problems characterized by non-convex and non-smooth loss functions, that are prevalent in FL applications but challenging to handle due to their intricate non-convexity and non-smoothness nature and the conflicting requirements on communication efficiency and privacy protection. In this paper, we propose a novel federated primal-dual algorithm with bidirectional model sparsification tailored for non-convex and non-smooth FL problems, and differential privacy is applied for privacy guarantee. Its unique insightful properties and some privacy and convergence analyses are also presented as the FL algorithm design guidelines. Extensive experiments on real-world data are conducted to demonstrate the effectiveness of the proposed algorithm and much superior performance than some state-of-the-art FL algorithms, together with the validation of all the analytical results and properties.