针对联邦图学习中的标签噪声问题，提出双视角过滤与自适应聚合方法。
FedRGL: Robust Federated Graph Learning for Label Noise
- 双视角一致性过滤：结合全局模型与子图结构动态阈值
- 引入图对比学习，提升编码器鲁棒性并生成高置信伪标签
- 基于未标记节点预测熵自适应聚合，适合有噪声的分布式图学习场景
联邦图学习（FGL）是一种基于图神经网络的分布式机器学习范式，支持客户端间安全协作建模本地图数据。然而，标签噪声会降低全局模型的泛化性能。现有联邦标签噪声学习方法主要面向计算机视觉，在FGL中表现不佳。为此，本文提出鲁棒联邦图学习方法FedRGL，引入双视角一致性噪声节点过滤机制，利用类感知动态阈值结合全局模型与子图结构。为增强客户端训练，引入图对比学习，提升编码器鲁棒性并为噪声节点赋予高置信伪标签。同时，通过未标记节点预测熵衡量模型质量，实现全局模型的自适应鲁棒聚合。在多个真实世界图数据集上的对比实验表明，FedRGL在不同噪声率、噪声类型及客户端数量下均优于12种基线方法。
原文摘要 · Abstract (English)
Federated Graph Learning (FGL) is a distributed machine learning paradigm based on graph neural networks, enabling secure and collaborative modeling of local graph data among clients. However, label noise can degrade the global model's generalization performance. Existing federated label noise learning methods, primarily focused on computer vision, often yield suboptimal results when applied to FGL. To address this, we propose a robust federated graph learning method with label noise, termed FedRGL. FedRGL introduces dual-perspective consistency noise node filtering, leveraging both the global model and subgraph structure under class-aware dynamic thresholds. To enhance client-side training, we incorporate graph contrastive learning, which improves encoder robustness and assigns high-confidence pseudo-labels to noisy nodes. Additionally, we measure model quality via predictive entropy of unlabeled nodes, enabling adaptive robust aggregation of the global model. Comparative experiments on multiple real-world graph datasets show that FedRGL outperforms 12 baseline methods across various noise rates, types, and numbers of clients.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。