标题： LEO-Split：一种在LEO卫星网络上的半监督分割学习框架 显示英文标题

标题： LEO-Split: A Semi-Supervised Split Learning Framework over LEO Satellite Networks

摘要： 最近，低轨卫星系统的广泛部署使得各种空间分析（例如，作物和气候监测）成为可能，这高度依赖于深度学习（DL）的进步。 然而，低轨卫星与地面站（GS）之间的间歇性连接显著阻碍了原始数据及时传输到GS进行集中式学习，而扩展的DL模型则妨碍了在资源受限的低轨卫星上的分布式学习。 尽管分割学习（SL）可以通过分割模型并将主要训练工作负载卸载到GS，从而成为解决这些问题的潜在方案，但劳动密集型的标记过程仍然是一个障碍，间歇性连接和数据异质性是其他挑战。 在本文中，我们提出了一种针对卫星网络的半监督（SS）SL设计，称为LEO-Split，以应对这些挑战。 利用半监督学习来处理（标记的）数据稀缺问题，我们构建了一个辅助模型来解决卫星-GS非接触时间的训练失败问题。 此外，我们提出了一种伪标记算法来校正卫星间的数据不平衡问题。 最后，设计了一种自适应激活插值方案，以防止GS端服务器子模型训练的过拟合问题。 使用真实世界低轨卫星轨迹（例如，Starlink）进行的大量实验表明，我们的LEO-Split框架相比最先进的基准表现出更优越的性能。 显示英文摘要

摘要： Recently, the increasing deployment of LEO satellite systems has enabled various space analytics (e.g., crop and climate monitoring), which heavily relies on the advancements in deep learning (DL). However, the intermittent connectivity between LEO satellites and ground station (GS) significantly hinders the timely transmission of raw data to GS for centralized learning, while the scaled-up DL models hamper distributed learning on resource-constrained LEO satellites. Though split learning (SL) can be a potential solution to these problems by partitioning a model and offloading primary training workload to GS, the labor-intensive labeling process remains an obstacle, with intermittent connectivity and data heterogeneity being other challenges. In this paper, we propose LEO-Split, a semi-supervised (SS) SL design tailored for satellite networks to combat these challenges. Leveraging SS learning to handle (labeled) data scarcity, we construct an auxiliary model to tackle the training failure of the satellite-GS non-contact time. Moreover, we propose a pseudo-labeling algorithm to rectify data imbalances across satellites. Lastly, an adaptive activation interpolation scheme is devised to prevent the overfitting of server-side sub-model training at GS. Extensive experiments with real-world LEO satellite traces (e.g., Starlink) demonstrate that our LEO-Split framework achieves superior performance compared to state-ofthe-art benchmarks.