arXiv cs.AI· Youming Tao, Weijun Yin, Huanyi Xie, Yanqing Yao, Dongxiao Yu, Xiuzhen Cheng, Di Wang, Falko Dressler·· 3 小时前AI 评分36
F-SOSA：面向边缘设备动态的一次性联邦模型编辑算法
One-Shot Federated Model Editing for Device Dynamics at the Edge
AI 导读
研究者提出 F-SOSA，首个支持快速精确模型编辑的稳定一次性联邦学习算法，通过本地模型子采样与曲率感知模型融合实现稳定聚合。在其基础上开发的 F-SOSA-U（遗忘）与 F-SOSA-I（增量学习）可在单轮通信内编辑模型，统计上与重新训练不可区分，仅花费预期 O(ρ) 比例的代价。城市移动仿真实验显示，其性能与重新训练相当，累计时间与通信开销分别至少降低 36% 和 22%。
正文 · 原文
Abstract:Standard federated learning (FL) requires persistent device connectivity across multiple communication rounds, making it unsuitable for dynamic environments such as vehicular networks, where devices frequently join and leave. One-shot FL offers a promising alternative by completing training in a single communication round. However, one-shot FL suffers from model solidification and cannot readily adapt to subsequent client dynamics, including new client contributions and the withdrawal of previous contributions. We therefore consider exact model editing, which removes withdrawn contributions through unlearning or incorporates newly contributed ones through incremental learning without retraining from scratch. We identify that efficient editing fundamentally hinges on learning stability. To this end, we propose F-SOSA, the first stable one-shot FL algorithm that enables rapid and exact model editing. F-SOSA leverages sub-sampling of local models with carefully tuned rate for stable aggregation and performs curvature-aware model fusion to preserve utility. We prove that F-SOSA achieves vanishing training error and reasonably small generalization error bounds. Building on this stability, we further develop two single-round editing algorithms: F-SOSA-U for unlearning and F-SOSA-I for incremental learning. They ensure that the edited model, along with all intermediate algorithmic states, is statistically indistinguishable from retraining while incurring only an expected $O(\rho)$ fraction of its cost, where $\rho$ quantifies stability. Experiments demonstrate that our approach matches retraining performance while reducing cumulative time and communication overhead by at least 36% and 22%, respectively, under realistic urban mobility simulations.
| Comments: | 30 pages, 7 figures. Accepted at IEEE Transactions on Networking |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Networking and Internet Architecture (cs.NI) |
| Cite as: | arXiv:2610.35212 [cs.LG] |
| (or arXiv:2610.35212v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35212 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Weijun Yin [view email]
[v1]
Mon, 26 Oct 2026 11:04:23 UTC (2,218 KB)
来源：arXiv cs.AI · arxiv.org