FedHST: Federated Hybrid Spatial–Temporal Optimization for robust non-IID learning
PLOS One · 12.11.2026 · 21 (11) · e0360546
DOI: 10.1371/journal.pone.0360546 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 12.11.2026. Verfasst von Duy-Dong Le und 1 weiteren Autor:innen. Beteiligte Einrichtung: University of Economics Ho Chi Minh City (UEH), Ho Chi Minh City, Vietnam. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: UEH.
Abstract (Originalsprache)
Server-side adaptive optimizers in Federated Learning (FL), such as FedAdam and FedAvgM, scale global updates using historical temporal statistics but do not explicitly model instantaneous cross-client variance within a single communication round. Under severe statistical heterogeneity (non-IID data), this can leave the adaptive step size insufficiently damped precisely when local client updates conflict most. We propose FedHST (Federated Hybrid Spatial–Temporal Optimization), a server-side aggregation framework that augments the temporal second-moment estimate with an instantaneous, coordinate-wise cross-client spatial variance term. This term acts as an additional damper under high client disagreement and vanishes under consensus, at which point FedHST reduces to standard temporal-only adaptation. We establish a non-convex convergence rate of 𝒪 ( 1 / T ) , with the residual term showing explicitly how the spatial damping coefficient γ moderates the effect of statistical heterogeneity on convergence. We evaluate FedHST on FashionMNIST, CIFAR-10, and CIFAR-100 under Dirichlet-partitioned non-IID splits ( α ∈ { 0.1 , 0.3 } ), comparing against FedAdam and FedDyn — the strongest representatives of the adaptive server-optimization and dynamic client-regularization paradigms, respectively — over three independent random seeds with paired significance testing; FedAvgM, FedYogi, FedAdagrad, FedProx, and SCAFFOLD are reviewed in Related Work. FedHST attains accuracy statistically indistinguishable from, or modestly and significantly exceeding (up to + 0.5 percentage points, p < 0.05 on FashionMNIST), FedAdam, while substantially and significantly outperforming FedDyn (+ 5 to + 28 percentage points across datasets, p < 0.01), at a computational overhead within 1–4% of FedAdam. Ablation studies further show that FedHST’s advantage over FedAdam grows systematically with the number of local training epochs and with non-IID severity, consistent with the mechanism by which spatial damping is designed to activate under cross-client conflict.
Autor:innen und Institutionen
- University of Economics Ho Chi Minh City (UEH), Ho Chi Minh City, Vietnam.
ORCID-Profile: Duy-Dong Le
Förderung
- UEH
Zitieren
APA
Le, D. D., & Dang, T. T. (2026). FedHST: Federated Hybrid Spatial–Temporal Optimization for robust non-IID learning. PLOS One, 21(11), e0360546. https://doi.org/10.1371/journal.pone.0360546
Vancouver
Le DD, Dang TT. FedHST: Federated Hybrid Spatial–Temporal Optimization for robust non-IID learning. PLOS One. 2026;21(11):e0360546. doi:10.1371/journal.pone.0360546
BibTeX
@article{Le2026,
author = {Le, Duy-Dong and Dang, Thai-Thinh},
title = {FedHST: Federated Hybrid Spatial–Temporal Optimization for robust non-IID learning},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {11},
pages = {e0360546},
doi = {10.1371/journal.pone.0360546}
}
EUVIMED