无需验证数据，通过服务器参数自动判断联邦学习何时停止。
Beyond Fixed Rounds: Data-Free Early Stopping for Practical Federated Learning
- 仅用服务器端参数监控任务向量增长速率来决定停止时机。
- 在皮肤病变等三类病理分类任务上，性能比基于验证数据的早停高12.3%以上。
- 只需不到3%的固定轮次预算即可筛选出差的配置，适合资源受限场景。
联邦学习（FL）支持去中心化协作学习而无需传输原始数据。然而，依赖固定全局轮次或验证数据进行超参数调优会带来高昂计算成本和隐私风险，阻碍实际部署。为此，我们提出一种无需数据的早停框架，通过仅监测服务器端参数的任务向量增长速率来确定最优停止点。在皮肤病变、血细胞和结肠病理分类任务上的数值结果表明，该方法在多种先进联邦学习方法中表现与基于验证数据的早停相当。具体而言，该框架平均需额外45/12/31轮即可实现比验证数据早停高出12.3%/8.9%/3.9%的性能提升；同时，仅需9/8/14轮额外开销即可完成不良配置筛选，低于固定轮次预算的3%。据我们所知，这是首个针对联邦学习提出的无数据早停框架。代码已开源。
原文摘要 · Abstract (English)
Federated Learning (FL) facilitates decentralized collaborative learning without transmitting raw data. However, reliance on fixed global rounds or validation data for hyperparameter tuning hinders practical deployment by incurring high computational costs and privacy risks. To address this, we propose a data-free early stopping framework that determines the optimal stopping point by monitoring the task vector's growth rate using only server-side parameters. The numerical results on skin lesion/blood cell/colon pathology classification demonstrate that our approach is comparable to the validation-based early stopping across various state-of-the-art FL methods. In particular, the proposed framework requires an average of 45/12/31 (skin lesion/blood cell/colon pathology) additional rounds to achieve over 12.3%/8.9%/3.9% higher performance than early stopping based on validation data. Moreover, the proposed framework requires only 9/8/14 additional rounds to screen bad configurations, which is less than 3% of the fixed-round budget. To the best of our knowledge, this is the first work to propose a data-free early stopping framework for FL methods. Our code is available at this open repository.
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。