Non-convex composite federated learning with heterogeneous data
This addresses communication efficiency and client drift issues in federated learning for distributed systems, representing an incremental improvement over existing methods.
The paper tackles the problem of federated learning with heterogeneous data by proposing an algorithm that decouples proximal operator evaluation and communication, using local updates to reduce communication frequency and client drift. It achieves sublinear and linear convergence to a bounded residual error under non-convexity and the proximal Polyak-Lojasiewicz inequality, demonstrating superiority over state-of-the-art methods in experiments.
We propose an innovative algorithm for non-convex composite federated learning that decouples the proximal operator evaluation and the communication between server and clients. Moreover, each client uses local updates to communicate less frequently with the server, sends only a single d-dimensional vector per communication round, and overcomes issues with client drift. In the analysis, challenges arise from the use of decoupling strategies and local updates in the algorithm, as well as from the non-convex and non-smooth nature of the problem. We establish sublinear and linear convergence to a bounded residual error under general non-convexity and the proximal Polyak-Lojasiewicz inequality, respectively. In the numerical experiments, we demonstrate the superiority of our algorithm over state-of-the-art methods on both synthetic and real datasets.