Optimizing Server Placement for Vertical Federated Learning in Dynamic Edge/Fog Networks
IEEE Transactions on Networking · 2026 · 34 · 5846-5866
DOI: 10.1109/ton.2026.3700898 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Networking, 2026. Verfasst von Su Wang und 2 weiteren Autor:innen. Die Arbeit nennt 80 Literaturangaben. Gefördert von: Innovation Grant from Princeton NextG.
Abstract (Originalsprache)
We investigate the control and optimization of vertical federated learning (VFL), a class of distributed machine learning (ML) methods in which edge/fog devices contain separate data features, in dynamic edge/fog networks. Owing to heterogeneous data features and hardware across edge/fog networks, devices’ contributions to VFL vary substantially, and, moreover, dynamic edge/fog networks can lead to the permanent exit or entry of select data features. In this setting, our proposed methodology, server controlled VFL in dynamic networks (SC-DN), first establishes the existence of a global first-order stationary point for every global round, and then leverages this result to jointly optimize ML model training and resource consumption based on four key control variables: 1) server placement; 2) device-to-server transmit power; 3) local device processor frequency; and 4) local training iterations per global round. The resulting optimization formulation contains coupled variables as well as numerous forms of logarithmic constraints which we show is a mixed-integer signomial program, an NP-hard problem, and for which we develop a general solver. Finally, via experiments on both image and multi-modal datasets, we show that our methodology demonstrates superior classification/regression performance and resource consumption savings than even greedy methodologies.
Förderung
- Innovation Grant from Princeton NextG
Zitieren
APA
Wang, S., Chiang, M., & Poor, H. V. (2026). Optimizing Server Placement for Vertical Federated Learning in Dynamic Edge/Fog Networks. IEEE Transactions on Networking, 34, 5846-5866. https://doi.org/10.1109/ton.2026.3700898
Vancouver
Wang S, Chiang M, Poor HV. Optimizing Server Placement for Vertical Federated Learning in Dynamic Edge/Fog Networks. IEEE Transactions on Networking. 2026;34:5846-5866. doi:10.1109/ton.2026.3700898
BibTeX
@article{Wang2026,
author = {Wang, Su and Chiang, Mung and Poor, H. Vincent},
title = {Optimizing Server Placement for Vertical Federated Learning in Dynamic Edge/Fog Networks},
journal = {IEEE Transactions on Networking},
year = {2026},
volume = {34},
pages = {5846--5866},
doi = {10.1109/ton.2026.3700898}
}
EUVIMED