FedRAW：在异步联邦学习中保留稀有标签影响力
FedRAW: Preserving Rare-Label Influence in Asynchronous Federated Learning
查看机构详情
- School of Artificial Intelligence and Data Science, Indian Institute of Technology Jodhpur（印度焦特布尔印度理工学院人工智能与数据科学学院）
- Department of Computer Science and Engineering, Indian Institute of Technology Jodhpur（印度焦特布尔印度理工学院计算机科学与工程系）
- Laboratoire d’Informatique de Grenoble (LIG), University of Grenoble Alpes（格勒诺布尔阿尔卑斯大学格勒诺布尔计算机实验室）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
FedRAW提出一种服务器端聚合方法，通过客户端更新去重和稀有标签感知加权，解决异步联邦学习中稀有标签客户端影响力不足导致的静默失败，提升稀有标签准确率并保持收敛。
中文摘要 AI 辅助
异步联邦学习通过从服务器端的客户端更新缓冲区中按到达顺序更新全局模型，而非等待所有选定客户端完成，从而提高了可扩展性。虽然这种方法高效，但在异构参与下，这种基于到达的聚合可能会悄然扭曲表示学习。我们识别出“静默稀有性失败”，这是一种隐藏的失败模式，其中持有稀有标签的客户端对全局模型的贡献过弱，即使其整体准确率看似基本不受影响。该失败源于两个耦合效应：稀有标签客户端在速度较慢或可用性较低时可能提交更新的频率较低，从而产生参与偏差；一旦其更新进入缓冲区，标准异步聚合不会赋予其补偿性影响力，从而产生聚合偏差。我们提出FedRAW，一种完全在服务器端的聚合方法，在不改变本地训练、客户端目标或通信协议的情况下保留稀有标签影响力。FedRAW结合了客户端级更新去重（防止频繁到达的客户端反复主导更新缓冲区）与稀有标签感知加权（增加携带低覆盖率标签的客户端的影响力）。我们通过参与偏差和聚合偏差形式化了静默稀有性失败，并证明FedRAW在保留收敛性的同时，相较于统一聚合增加了稀有标签客户端的影响力。在EMNIST Balanced、CIFAR-10、HAM10000和ISIC-2019上，FedRAW提高了稀有标签准确率，同时保持了相当的全局准确率，并增加了可忽略的服务器端计算开销。
英文摘要
Asynchronous federated learning improves scalability by updating the global model from a server-side buffer of client updates as they arrive, rather than waiting for all selected clients to finish. While efficient, this arrival-driven aggregation can silently distort representation learning under heterogeneous participation. We identify silent rarity failure, a hidden failure mode in which clients holding rare labels contribute too weakly to the global model even though its overall accuracy appears largely unaffected. This failure arises from two coupled effects: rare-label clients may submit updates less frequently when they are slower or less available, creating participation bias; and once their updates enter the buffer, standard asynchronous aggregation assigns them no compensating influence, creating aggregation bias. We propose FedRAW, a fully server-side aggregation method that preserves rare-label influence without changing local training, client objectives, or communication protocols. FedRAW combines client-level update deduplication, which prevents frequently arriving clients from repeatedly dominating the update buffer, with rare-label-aware weighting, which increases the influence of clients carrying low-coverage labels. We formalize silent rarity failure through participation and aggregation bias, and show that FedRAW increases rare-label client influence over uniform aggregation while preserving convergence. Across EMNIST Balanced, CIFAR-10, HAM10000, and ISIC-2019, FedRAW improves rarelabel accuracy while preserving comparable global accuracy and adding negligible server-side computation.
