arXiv：cs.LG· Seungjun Lee, Ensieh Khazaei, Dimitrios Hatzinakos, Baturalp Buyukates, Sunwoo Lee·· 2 天前AI 评分33
通过潜在信息共享加速联邦学习
Latent Information Sharing for Accelerating Federated Learning
AI 导读
研究提出一种潜在信息共享方案，通过共享少量隐藏层激活直接缓解联邦学习中的数据异构问题，在保持收敛保证与数据隐私的同时显著提升训练效率。与 FedProx、SCAFFOLD、FedPVR、FedProto、SplitFed 等方法相比，该方法在固定轮次预算下取得更优模型精度，且未带来过多通信开销。
正文
Abstract:Federated learning (FL) is a communication-efficient distributed learning paradigm. However, client drift remains one of the most critical challenges, hindering the efficient training of a global model. In this study, we propose a novel latent information sharing scheme that directly mitigates data heterogeneity across clients. Our theoretical and empirical results show that sharing a small amount of hidden-layer activations significantly improves training efficiency while preserving convergence guarantees and data privacy. Furthermore, we compare our method with existing FL approaches designed to address client drift, including FedProx, SCAFFOLD, FedPVR, FedProto, and SplitFed, and demonstrate superior model accuracy under a fixed round budget without incurring excessive communication overhead. Overall, this work presents a promising new knowledge aggregation scheme and provides a comprehensive analysis of the impact of activation sharing on federated optimization.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.01126 [cs.LG] |
| (or arXiv:2610.01126v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.01126 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Seungjun Lee [view email]
[v1]
Thu, 1 Oct 2026 06:11:14 UTC (638 KB)
来源：arXiv：cs.LG · arxiv.org