arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2405.01142cs.LG
收藏

异构数据上顺序联邦学习的尖锐界

Sharp Bounds for Sequential Federated Learning on Heterogeneous Data

National Engineering Research Center for Mobile Network Technologies Beijing University of Posts and Telecommunications(移动网络技术国家工程研究中心 北京邮电大学)

Yipeng Li, Xinchen Lyu

AI总结:

针对异构数据上的顺序联邦学习,建立上下界匹配的尖锐收敛保证,证明其优于并行联邦学习,并通过实验验证反直觉结论。

摘要:

联邦学习(FL)中存在两种范式:并行联邦学习(PFL),其中模型在客户端之间以并行方式训练;以及顺序联邦学习(SFL),其中模型在客户端之间以顺序方式训练。具体而言,在PFL中,客户端独立执行本地更新并将更新后的模型参数发送给全局服务器进行聚合;而在SFL中,一个客户端只有在接收到序列中前一个客户端的模型参数后才开始其本地更新。与PFL相比,SFL在异构数据上的收敛理论仍然缺乏。为了解决SFL的理论困境,我们为异构数据上的SFL建立了具有上界和下界的尖锐收敛保证。具体而言,我们推导了强凸、一般凸和非凸目标函数的上界,并为强凸和一般凸目标函数构造了匹配的下界。然后,我们将SFL的上界与PFL的上界进行比较,表明在异构数据上SFL优于PFL(至少当异构程度相对较高时)。实验结果验证了这一反直觉的理论发现。

英文摘要:

There are two paradigms in Federated Learning (FL): parallel FL (PFL), where models are trained in a parallel manner across clients, and sequential FL (SFL), where models are trained in a sequential manner across clients. Specifically, in PFL, clients perform local updates independently and send the updated model parameters to a global server for aggregation; in SFL, one client starts its local updates only after receiving the model parameters from the previous client in the sequence. In contrast to that of PFL, the convergence theory of SFL on heterogeneous data is still lacking. To resolve the theoretical dilemma of SFL, we establish sharp convergence guarantees for SFL on heterogeneous data with both upper and lower bounds. Specifically, we derive the upper bounds for the strongly convex, general convex and non-convex objective functions, and construct the matching lower bounds for the strongly convex and general convex objective functions. Then, we compare the upper bounds of SFL with those of PFL, showing that SFL outperforms PFL on heterogeneous data (at least, when the level of heterogeneity is relatively high). Experimental results validate the counterintuitive theoretical finding.

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。

↑