针对卫星数据非独立同分布问题，提出自适应个性化联邦学习方法
FedOrbit: Adaptive Personalized Federated Learning for Non-IID LEO Satellite Constellations
- 基于轨道级类别感知的分层聚合与动态特征分解
- 在六组实验中五次达最高准确率，超越最强基线超8.6个百分点
- 适合处理低轨卫星星座中数据分布不均的场景
低轨卫星星座中的联邦学习受非独立同分布数据和地面站可见性不规则影响，二者均由轨道几何决定。当轨道级类别分布不重叠时，全局聚合效果差；而当分布重叠时，强个性化又可能过度。我们提出FedOrbit，结合跨星间链路的连续轨道级训练、类别感知的分层聚合、基于返回率抑制的质量加权特征聚合，以及基于轨道间类别相似性的自适应特征分解。在三个遥感基准和两个非独立同分布划分下，FedOrbit在六组设置中五次达到最高准确率，第六次仅比最优结果低0.9个百分点。相较于最强基线，在狄利克雷划分下提升16.1个百分点，在病态划分下提升8.6个百分点，且在五组设置中轨道间准确率差异最小。
原文摘要 · Abstract (English)
Federated learning (FL) in Low Earth Orbit (LEO) satellite constellations is affected by non-IID data and irregular ground-station visibility, both driven by orbital geometry. Global aggregation performs poorly when orbit-level class distributions are disjoint, while strong personalisation can be excessive when these distributions overlap. We present FedOrbit, which combines continuous orbit-level training over inter-satellite links, class-aware hierarchical aggregation, quality-weighted feature aggregation with return-rate dampening, and adaptive feature decomposition based on inter-orbit class similarity. Across three remote-sensing benchmarks and two non-IID partitions, FedOrbit achieves the highest accuracy in five of six settings and is within $0.9$ percentage points of the best result in the sixth. The gains over the strongest baseline reach $16.1$ percentage points under Dirichlet partitioning and $8.6$ under pathological partitioning, with the smallest per-orbit accuracy spread in five of six settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。