English

Gradient Tracking for High Dimensional Federated Optimization

Optimization and Control 2023-12-12 v1 Methodology

Abstract

In this paper, we study the (decentralized) distributed optimization problem with high-dimensional sparse structure. Building upon the FedDA algorithm, we propose a (Decentralized) FedDA-GT algorithm, which combines the \textbf{gradient tracking} technique. It is able to eliminate the heterogeneity among different clients' objective functions while ensuring a dimension-free convergence rate. Compared to the vanilla FedDA approach, (D)FedDA-GT can significantly reduce the communication complexity, from O(s2logd/ε3/2){O}(s^2\log d/\varepsilon^{3/2}) to a more efficient O(s2logd/ε){O}(s^2\log d/\varepsilon). In cases where strong convexity is applicable, we introduce a multistep mechanism resulting in the Multistep ReFedDA-GT algorithm, a minor modified version of FedDA-GT. This approach achieves an impressive communication complexity of O(slogdlog1ε){O}\left(s\log d \log \frac{1}{\varepsilon}\right) through repeated calls to the ReFedDA-GT algorithm. Finally, we conduct numerical experiments, illustrating that our proposed algorithms enjoy the dual advantage of being dimension-free and heterogeneity-free.

Keywords

Cite

@article{arxiv.2312.05590,
  title  = {Gradient Tracking for High Dimensional Federated Optimization},
  author = {Jiadong Liang and Yang Peng and Zhihua Zhang},
  journal= {arXiv preprint arXiv:2312.05590},
  year   = {2023}
}
R2 v1 2026-06-28T13:45:54.291Z