KoReA-SFL：基于知识回放的拆分联邦学习以对抗灾难性遗忘
KoReA-SFL: Knowledge Replay-based Split Federated Learning Against Catastrophic Forgetting
AI总结：
针对拆分联邦学习忽视数据异构性和灾难性遗忘导致精度低的问题，提出KoReA-SFL，采用多模型聚合和知识回放策略，在非IID和IID场景下测试精度最高提升23.25%。
摘要：
尽管拆分联邦学习（SFL）擅长在资源受限的客户端之间实现知识共享，但由于忽视了数据异构性和灾难性遗忘问题，其训练精度较低。为解决这一问题，我们提出了一种新颖的SFL方法，名为KoReA-SFL，它采用多模型聚合机制来缓解由异构数据引起的梯度分歧，并采用知识回放策略来应对灾难性遗忘。具体而言，在KoReA-SFL中，云服务器（即联邦服务器和主服务器）维护多个分支模型部分而非一个全局部分用于本地训练，并维护一个聚合的主模型部分用于分支部分之间的知识共享。为避免灾难性遗忘，KoReA-SFL的主服务器根据每个服务器端分支模型部分的训练数据分布，选择多个辅助设备进行知识回放。在非独立同分布（non-IID）和独立同分布（IID）场景下获得的实验结果表明，KoReA-SFL显著优于传统的SFL方法（测试精度最高提升23.25%）。
英文摘要：
Although Split Federated Learning (SFL) is good at enabling knowledge sharing among resource-constrained clients, it suffers from the problem of low training accuracy due to the neglect of data heterogeneity and catastrophic forgetting. To address this issue, we propose a novel SFL approach named KoReA-SFL, which adopts a multi-model aggregation mechanism to alleviate gradient divergence caused by heterogeneous data and a knowledge replay strategy to deal with catastrophic forgetting. Specifically, in KoReA-SFL cloud servers (i.e., fed server and main server) maintain multiple branch model portions rather than a global portion for local training and an aggregated master-model portion for knowledge sharing among branch portions. To avoid catastrophic forgetting, the main server of KoReA-SFL selects multiple assistant devices for knowledge replay according to the training data distribution of each server-side branch-model portion. Experimental results obtained from non-IID and IID scenarios demonstrate that KoReA-SFL significantly outperforms conventional SFL methods (by up to 23.25\% test accuracy improvement).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
