arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.17876cs.LGcs.DCmath.OC
收藏

去中心化定向协作用于个性化联邦学习

Decentralized Directed Collaboration for Personalized Federated Learning

Nanjing University of Science and Technology(南京理工大学);Tsinghua University(清华大学);National University of Defense Technology(国防科技大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));JD Explore Academy(京东探索研究院)

Yingqi Liu, Yifan Shi, Qinglun Li, Baoyuan Wu, Xueqian Wang, Li Shen

AI总结:

针对去中心化个性化联邦学习中的异构性问题,提出定向协作框架DFedPGP,结合随机梯度推送与部分模型个性化,实现更优收敛和SOTA精度。

摘要:

个性化联邦学习(PFL)旨在为每个客户端找到最优的个性化模型。为避免基于服务器的联邦学习中存在的中心故障和通信瓶颈问题,我们专注于去中心化个性化联邦学习(DPFL),它以点对点(P2P)方式进行分布式模型训练。目前DPFL中的大多数个性化工作基于无向且对称的拓扑结构,然而,数据、计算和通信资源的异构性导致个性化模型存在较大差异,这使得无向聚合达到次优的个性化性能且无法保证收敛。为解决这些问题,我们提出了一种结合随机梯度推送和部分模型个性化的定向协作DPFL框架,称为去中心化联邦部分梯度推送(DFedPGP)。该方法对现代深度模型中的线性分类器进行个性化以定制局部解,并以完全去中心化的方式学习共识表示。客户端仅基于定向且非对称的拓扑结构与部分邻居共享梯度,这保证了资源效率的灵活选择并带来更好的收敛性。理论上,我们证明了所提出的DFedPGP在一般非凸设置下实现了优越的收敛速率$\backslash mathcal{O}(\backslash frac{1}{\backslash sqrt{T}})$,并证明了客户端之间更紧密的连接会加速收敛。所提出的方法在数据和计算异构场景下均达到了最先进的准确率,验证了定向协作和部分梯度推送的有效性。

英文摘要:

Personalized Federated Learning (PFL) is proposed to find the greatest personalized models for each client. To avoid the central failure and communication bottleneck in the server-based FL, we concentrate on the Decentralized Personalized Federated Learning (DPFL) that performs distributed model training in a Peer-to-Peer (P2P) manner. Most personalized works in DPFL are based on undirected and symmetric topologies, however, the data, computation and communication resources heterogeneity result in large variances in the personalized models, which lead the undirected aggregation to suboptimal personalized performance and unguaranteed convergence. To address these issues, we propose a directed collaboration DPFL framework by incorporating stochastic gradient push and partial model personalized, called \textbf{D}ecentralized \textbf{Fed}erated \textbf{P}artial \textbf{G}radient \textbf{P}ush (\textbf{DFedPGP}). It personalizes the linear classifier in the modern deep model to customize the local solution and learns a consensus representation in a fully decentralized manner. Clients only share gradients with a subset of neighbors based on the directed and asymmetric topologies, which guarantees flexible choices for resource efficiency and better convergence. Theoretically, we show that the proposed DFedPGP achieves a superior convergence rate of $\mathcal{O}(\frac{1}{\sqrt{T}})$ in the general non-convex setting, and prove the tighter connectivity among clients will speed up the convergence. The proposed method achieves state-of-the-art (SOTA) accuracy in both data and computation heterogeneity scenarios, demonstrating the efficiency of the directed collaboration and partial gradient push.

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。

↑