Decentralized machine learning has broadened its scope recently with the invention of Federated Learning (FL), Split Learning (SL), and their hybrids like Split Federated Learning (SplitFed or SFL). The goal of SFL is to reduce the computational power required by each client in FL and parallelize SL while maintaining privacy. This paper investigates the robustness of SFL against packet loss on communication links. The performance of various SFL aggregation strategies is examined by splitting the model at two points -- shallow split and deep split -- and testing whether the split point makes a statistically significant difference to the accuracy of the final model. Experiments are carried out on a segmentation model for human embryo images and indicate the statistically significant advantage of a deeper split point.

翻译：近年来，随着联邦学习（FL）、分割学习（SL）及其混合方法如分割联邦学习（SplitFed或SFL）的出现，去中心化机器学习拓展了其应用范围。SFL的目标是在降低FL中每个客户端所需计算能力的同时实现SL的并行化并保护隐私。本文研究了SFL在通信链路上对数据包丢失的鲁棒性。通过将模型在两个点（浅层分割和深层分割）进行分割，并检验分割点是否对最终模型的准确性产生统计上的显著差异，评估了多种SFL聚合策略的性能。实验基于人类胚胎图像的分割模型进行，结果表明更深的模型分割点在统计上具有显著优势。