DecFedFreq: Frequency‐Guided Cluster‐Restricted Encoder Residual Aggregation for Decoupled Personalized Federated Learning at the Edge
Concurrency and Computation: Practice and Experience · 2026-11 · 38 (22)
DOI: 10.1002/cpe.71029 ↗Kurz gefasst
Fachartikel in der Zeitschrift Concurrency and Computation: Practice and Experience, online erschienen am 08.11.2026. Verfasst von Wanqiang Song und 4 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Gefördert von: Natural Science Foundation of Anhui Province, National Natural Science Foundation of China.
Abstract (Originalsprache)
Heterogeneous edge‐intelligence systems require federated models that preserve useful shared representations while adapting to persistent client differences. Existing decoupled personalized federated learning methods separate a shared encoder from private prediction heads and coordinate encoder information across clients. This paper proposes DecFedFreq, a framework that makes encoder residual aggregation adaptive. The method uses a low‐frequency discrete cosine transform descriptor as a deterministic, compressed summary of encoder‐update residuals for similarity estimation, applies affinity‐based grouping to infer dynamic client clusters, and restricts complete encoder‐residual aggregation to those clusters while prediction heads remain local. Across 10 matched random seeds under severe heterogeneous partitions with full participation, DecFedFreq obtains final‐round accuracies of on Fashion‐MNIST, on CIFAR‐10, and on CIFAR‐100. Relative to FedRep, the corresponding paired gains are 1.33, 2.96, and 5.13 percentage points, respectively, with all two‐sided paired ‐tests yielding . Additional five‐seed Fashion‐MNIST analyses of heterogeneous availability, unequal local‐computation budgets, representation alternatives, and a single‐global‐cluster control indicate that the main measured gain arises from similarity‐based grouping and cluster‐restricted aggregation. Discrete Cosine Transform is therefore positioned as the deterministic, empirically stable default similarity representation in the tested setting, with competitive accuracy and an eightfold reduction in similarity‐input dimension relative to the complete raw residual, rather than as an independently optimal accuracy component; server‐only profiling quantifies the added computation and memory over 25–100 active clients.
Förderung
- Natural Science Foundation of Anhui Province
- National Natural Science Foundation of China
Zitieren
APA
Song, W., Shi, L., Guo, D., Wang, Z., & Wang, J. (2026). DecFedFreq: Frequency‐Guided Cluster‐Restricted Encoder Residual Aggregation for Decoupled Personalized Federated Learning at the Edge. Concurrency and Computation: Practice and Experience, 38(22). https://doi.org/10.1002/cpe.71029
Vancouver
Song W, Shi L, Guo D, Wang Z, Wang J. DecFedFreq: Frequency‐Guided Cluster‐Restricted Encoder Residual Aggregation for Decoupled Personalized Federated Learning at the Edge. Concurrency and Computation: Practice and Experience. 2026;38(22). doi:10.1002/cpe.71029
BibTeX
@article{Song2026,
author = {Song, Wanqiang and Shi, Lei and Guo, Dong and Wang, Zhengqi and Wang, Jiawei},
title = {DecFedFreq: Frequency‐Guided Cluster‐Restricted Encoder Residual Aggregation for Decoupled Personalized Federated Learning at the Edge},
journal = {Concurrency and Computation: Practice and Experience},
year = {2026},
volume = {38},
number = {22},
doi = {10.1002/cpe.71029}
}
EUVIMED