FedRG：释放表征几何以应对含噪声客户端的联邦学习
FedRG: Unleashing the Representation Geometry for Federated Learning with Noisy Clients
用自监督得到的标签无关球面表征几何（vMF 混合聚类）来判别噪声样本，替代仅依赖标量损失的思路；做噪声标签、长尾或数据异构下的鲁棒训练时，可把这套「表征几何优先」的样本筛选与软标签映射机制迁移到视觉分类流水线中。
原文依据 FedRG creates label-agnostic spherical representations by using self-supervision. It then iteratively fits a spherical von Mises-Fisher (vMF) mixture model
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论联邦学习（FL）因分布式场景中不可避免的噪声标注而出现性能退化。现有方法通过利用损失值从数据集中区分噪声样本以进行标签纠正，已取得一定进展。然而，在异构场景下，依赖标量损失的噪声样本识别对联邦学习而言缺乏可靠性。本文从表征视角重新审视这一范式，提出 FedRG（Fed erated under R epresentation G eometry），遵循「表征几何优先」原则来识别噪声标签。首先，FedRG 利用自监督构建与标签无关的球面表征；随后迭代地将球面 von Mises-Fisher（vMF）混合模型拟合到该几何结构上，以捕获语义簇。该几何证据与语义-标签软映射机制相结合，在无标签特征空间与带标注标签条件特征空间之间导出分布散度，从而稳健地识别噪声样本，并用新分离出的干净数据集更新语义-标签映射。最后，我们在噪声标签上引入额外的个性化噪声吸收矩阵，以实现稳健优化。大量实验结果表明，在多种噪声客户端场景下，FedRG 在数据异构的联邦学习中优于最优方法。
英文原文摘要 1427 字符
Federated learning (FL) suffers from performance degradation due to the inevitable presence of noisy annotations in distributed scenarios. Existing approaches have advanced in distinguishing noisy samples from the dataset for label correction by leveraging loss values. However, noisy samples recognition relying on scalar loss lacks reliability for FL under heterogeneous scenarios. In this paper, we rethink this paradigm from a representation perspective and propose FedRG(**Fed**erated under **R**epresentation **G**emometry), which follows **"the principle of "representation geometry priority"** to recognize noisy labels. Firstly, FedRG creates label-agnostic spherical representations by using self-supervision. It then iteratively fits a spherical von Mises-Fisher (vMF) mixture model to this geometry to capture semantic clusters. This geometric evidence is integrated with a semantic-label soft mapping mechanism to derive a distribution divergence between the label-free and annotated label-conditioned feature space, which robustly identifies noisy samples and updates the semantic-label mappings with the newly separated clean dataset. Lastly, we employ an additional personalized noise absorption matrix on noisy labels to achieve robust optimization. Extensive experimental results demonstrate that FedRG outperforms state-of-the-art methods for FL with data heterogeneity under diverse noisy clients scenarios.