Toward Communication-Efficient Decentralized Federated Graph Learning Over Non-IID Data
IEEE Transactions on Mobile Computing · 2026-05 · 25 (5) · 6929-6947
DOI: 10.1109/tmc.2025.3641696 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2026-05. Verfasst von Shilong Wang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 75 Literaturangaben.
Abstract (Originalsprache)
Decentralized Federated Graph Learning (DFGL) overcomes the potential bottlenecks of the parameter server in FGL. However, extensive cross-worker communication of graph node embeddings during DFGL training introduces substantial communication costs. To improve communication efficiency, constructing sparse network topologies or applying graph sampling are potential methods. In this paper, we first reveal the bidirectional coupling between network topology construction and graph sampling, underscoring the necessity of their joint optimization. Motivated by this insight, we propose Duplex , a unified framework that co-optimizes these two components by explicitly modeling their interdependent relationship, thereby significantly reducing communication costs while enhancing training performance in DFGL. Duplex formulates the decision-making process as a coordinated configuration $\langle \mathbf {A}, \mathbf {R} \rangle$ 〈 A , R 〉 , where $\mathbf {A}$ A is the adjacency matrix of the network topology and $\mathbf {R}$ R denotes the set of graph sampling ratios for workers. However, determining proper coordinated configurations to achieve optimal communication efficiency and training performance (e.g., model accuracy and convergence rate) is challenging due to several practical issues, e.g. , statistical heterogeneity and dynamic network conditions. To overcome these challenges, Duplex introduces a novel learning-driven algorithm to adaptively determine optimal network topologies and graph sampling ratios for workers. Experimental results demonstrate that Duplex reduces completion time by 20.1%–48.8% and communication costs by 16.7%–37.6% to achieve target accuracy, while improving accuracy by 3.3%–7.9% under identical resource budgets compared to baselines.
Zitieren
APA
Wang, S., Liu, J., Xu, H., Tang, C., Ma, Q., & Huang, L. (2026). Toward Communication-Efficient Decentralized Federated Graph Learning Over Non-IID Data. IEEE Transactions on Mobile Computing, 25(5), 6929-6947. https://doi.org/10.1109/tmc.2025.3641696
Vancouver
Wang S, Liu J, Xu H, Tang C, Ma Q, Huang L. Toward Communication-Efficient Decentralized Federated Graph Learning Over Non-IID Data. IEEE Transactions on Mobile Computing. 2026;25(5):6929-6947. doi:10.1109/tmc.2025.3641696
BibTeX
@article{Wang2026,
author = {Wang, Shilong and Liu, Jianchun and Xu, Hongli and Tang, Chenxia and Ma, Qianpiao and Huang, Liusheng},
title = {Toward Communication-Efficient Decentralized Federated Graph Learning Over Non-IID Data},
journal = {IEEE Transactions on Mobile Computing},
year = {2026},
volume = {25},
number = {5},
pages = {6929--6947},
doi = {10.1109/tmc.2025.3641696}
}
EUVIMED