Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedRGL: Robust Federated Graph Learning under Label Noise
De Li, Zhou Tan, Qiyu Li, Zeming Gan, Tiange Xia, Chunpei Li, Xianxian LI
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add FedRGL on top of FedAvg under Uniform noise | FedAvg · Cora (5 clients), Uniform noise 0.3 | accuracy: 48.19 -> 78.75 | Improved |
| add FedRGL on top of FedAvg under Pair noise | FedAvg · Cora (5 clients), Pair noise 0.3 | accuracy: 55.80 -> 75.14 | Improved |
| add FedRGL on top of FedAvg under Uniform noise | FedAvg · Photo (20 clients), Uniform noise 0.3 | accuracy: 72.55 -> 84.06 | Improved |
| use FedRGL instead of FedAvg with clean labels | FedAvg · PubMed (10 clients), clean labels | accuracy: 85.86 -> 85.33 | No clear effect |
| use FedRGL instead of CRGNN with clean labels | CRGNN · Cora (5 clients), clean labels | accuracy: 83.09 -> 82.12 | Hurt |
| use FedRGL instead of CRGNN under Uniform noise | CRGNN · CS (10 clients), Uniform noise 0.3 | accuracy: 81.30 -> 87.79 | Improved |
| add FedRGL as the number of noisy clients rises to 100% | FedAvg · Photo (20 clients), Pair noise 0.4, 100% noisy clients | accuracy: 53.4 -> 73.4 | Improved |
| use FedRGL with few noisy clients (20%) | FedAvg · Photo (20 clients), Pair noise 0.4, 20% noisy clients | accuracy: 85.7 -> 87.5 | Improved |
| apply FedRGL on large-scale graph with 20 clients | FedAvg · ogbn-arxiv (20 clients), Uniform noise 0.4 | accuracy: 51.83 -> 56.61 | Improved |
| apply FedRGL on large-scale graph with 30 clients, Pair noise | FedAvg · ogbn-arxiv (30 clients), Pair noise 0.4 | accuracy: 45.33 -> 50.83 | Improved |
| add the global model view to noisy node filtering (ablation) | FedRGL · Cora and CS, noise 0.3 | accuracy | Improved |
| add the local structure view to noisy node filtering (ablation) | FedRGL · Cora and CS, noise 0.3 | accuracy | Improved |
Closest papers by meaning
- FedDiv: Collaborative Noise Filtering for Federated Learning with Noisy Labels (AAAI 2024 · similarity 0.90)
- $S^2$FGL: Spatial Spectral Federated Graph Learning (ICML 2025 · similarity 0.90)
- FedClean: A General Robust Label Noise Correction for Federated Learning (ICML 2025 · similarity 0.90)
- Virtual Nodes Can Help: Tackling Distribution Shifts in Federated Graph Learning (AAAI 2025 · similarity 0.89)
- Federated Graph-Level Clustering Network (AAAI 2025 · similarity 0.89)
- Subgraph Federated Learning for Local Generalization (ICLR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
