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 2308.01296cs.DC
收藏

基于区块链的层次化联邦学习中的掉队者缓解与延迟优化

Straggler Mitigation and Latency Optimization in Blockchain-based Hierarchical Federated Learning

Zhilin Wang, Qin Hu, Minghui Xu, Zeihui Xiong

AI总结:

提出基于区块链的去中心化层次化联邦学习框架BHFL,利用Raft联盟区块链保障可信聚合,设计HieAvg聚合方法利用历史权重估计掉队者缺失提交,并联合优化模型收敛与共识延迟,在非凸和非IID数据下性能优于传统方法。

摘要:

云-边-端层次化联邦学习(HFL)最近被提出,以实现通信高效且保护隐私的分布式学习。然而,存在几个关键挑战,例如单点故障以及边缘服务器和本地设备中潜在的掉队者。为了解决这些问题,我们提出了一种去中心化且容忍掉队者的基于区块链的HFL(BHFL)框架。具体而言,在边缘服务器上部署基于Raft的联盟区块链,为BHFL中的全局模型聚合提供分布式且可信的计算环境。为了缓解掉队者对学习的影响,我们提出了一种新的聚合方法HieAvg,该方法利用掉队者的历史权重来估计缺失的提交。此外,我们通过联合考虑全局模型收敛和区块链共识延迟的约束,优化了BHFL的整体延迟。理论分析和实验评估表明,我们提出的基于HieAvg的BHFL可以在存在掉队者的情况下收敛,即使损失函数是非凸的且本地设备上的数据是非独立同分布的(non-IID),其性能也优于传统方法。

英文摘要:

Cloud-edge-device hierarchical federated learning (HFL) has been recently proposed to achieve communication-efficient and privacy-preserving distributed learning. However, there exist several critical challenges, such as the single point of failure and potential stragglers in both edge servers and local devices. To resolve these issues, we propose a decentralized and straggler-tolerant blockchain-based HFL (BHFL) framework. Specifically, a Raft-based consortium blockchain is deployed on edge servers to provide a distributed and trusted computing environment for global model aggregation in BHFL. To mitigate the influence of stragglers on learning, we propose a novel aggregation method, HieAvg, which utilizes the historical weights of stragglers to estimate the missing submissions. Furthermore, we optimize the overall latency of BHFL by jointly considering the constraints of global model convergence and blockchain consensus delay. Theoretical analysis and experimental evaluation show that our proposed BHFL based on HieAvg can converge in the presence of stragglers, which performs better than the traditional methods even when the loss function is non-convex and the data on local devices are non-independent and identically distributed (non-IID).

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。

↑