Budget-Prioritized Dynamic Regrouping for Edge Federated Services Under Workload Drift and Privacy-Budget Constraints
Li Zhao, Long Chen, Zhongyi Chen · 2026
Read Budget-Prioritized Dynamic Regrouping for Edge Federated Services Under Workload Drift and Privacy-Budget Constraints by Li Zhao, Long Chen, Zhongyi Chen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13605417.txt.
Description
Li Zhao, Long Chen, and Zhongyi Chen manage shifting groups in edge federated learning in a paper from two thousand twenty-six. Devices train a shared model without sending raw data, but fixed groups grow stale as workloads drift. Regrouping too often wastes effort and burns the privacy budget.
Their method regroups only when a trigger fires: one local move per time slot, accepted only if it stays within the privacy budget and beats a threshold covering move costs. They prove the trigger tightens as costs rise or budget shrinks. Trace-driven tests on a real cluster data set kept utility competitive while holding regrouping in check. A heart disease federated test confirmed the method plugs into group-aware training.
Catalog record
| CREATOR | Li Zhao, Long Chen, Zhongyi Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |