English

FedNS: A Fast Sketching Newton-Type Algorithm for Federated Learning

Machine Learning 2024-01-08 v1 Distributed, Parallel, and Cluster Computing

Abstract

Recent Newton-type federated learning algorithms have demonstrated linear convergence with respect to the communication rounds. However, communicating Hessian matrices is often unfeasible due to their quadratic communication complexity. In this paper, we introduce a novel approach to tackle this issue while still achieving fast convergence rates. Our proposed method, named as Federated Newton Sketch methods (FedNS), approximates the centralized Newton's method by communicating the sketched square-root Hessian instead of the exact Hessian. To enhance communication efficiency, we reduce the sketch size to match the effective dimension of the Hessian matrix. We provide convergence analysis based on statistical learning for the federated Newton sketch approaches. Specifically, our approaches reach super-linear convergence rates w.r.t. the communication rounds for the first time. We validate the effectiveness of our algorithms through various experiments, which coincide with our theoretical findings.

Keywords

Cite

@article{arxiv.2401.02734,
  title  = {FedNS: A Fast Sketching Newton-Type Algorithm for Federated Learning},
  author = {Jian Li and Yong Liu and Wei Wang and Haoran Wu and Weiping Wang},
  journal= {arXiv preprint arXiv:2401.02734},
  year   = {2024}
}

Comments

Accepted at AAAI 2024

R2 v1 2026-06-28T14:09:26.099Z