arXiv：cs.LG（机器学习，全量分类）· Wentao Yue, Tianyou Lai, Hongji Li, Qingyu Mao, Qilei Li·· 15 小时前AI 评分32
FedSAP：面向多域异构边缘设备的结构化自适应分区联邦学习框架
FedSAP: Federated Learning with Structured Adaptive Partitioning for Multi-Domain Heterogeneous Edge Devices
AI 导读
FedSAP 将结构化剪枝建模为预算约束下的三态通道分配，把通道分为 Global 池、伪域专属 Private 池和 Dropped 状态，并配合 Domain-Guided Assignment 与 Type-Matched Aggregation 限制各通道的共享范围。
正文
Abstract:Federated learning (FL) on heterogeneous edge devices must jointly accommodate unequal resource budgets and domain-shifted local data. Existing resource-adaptive methods decide how much of a model each client trains but not where retained capacity should reside or how it should be shared, whereas federated domain-generalization methods usually assume a shared full architecture. Uniform compression can therefore discard high-utility channels, and a single aggregation path can mix transferable features with domain-sensitive updates. We propose FedSAP, a domain-aware heterogeneous FL framework that casts structured pruning as budget-constrained tri-state channel allocation. FedSAP converts each keep ratio into non-uniform layer budgets, assigns stable channels to a Global pool, useful domain-sensitive channels to pseudo-domain-specific Private pools, and low-utility channels to a Dropped state. This partition lets broadly useful features benefit from cross-client pooling while isolating domain-sensitive updates from incompatible clients. Domain-Guided Assignment infers pseudo-domains from shallow-gradient similarity, while Type-Matched Aggregation restricts each channel to its intended sharing scope. Across three random seeds, FedSAP reaches 76.00% and 72.67% mean global accuracy on Digits and Office-Caltech, exceeding the strongest baseline by 1.70 and 4.92 percentage points while supporting client pruning ratios of up to 80% across heterogeneous clients.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.01638 [cs.LG] |
| (or arXiv:2610.01638v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.01638 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wentao Yue [view email]
[v1]
Thu, 1 Oct 2026 13:05:01 UTC (1,648 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org