Research map / Federated learning and privacy / Personalized and heterogeneous FL
Causality-inspired Federated Learning for Dynamic Spatio-Temporal Graphs
Yuxuan Liu, Wenchao Xu, Haozhao Wang, Zhiming He, Zhaofeng Shi, Chongyang Xu, Peichao Wang, Boyuan Zhang
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 |
|---|---|---|---|
| use SC-FSGL instead of FUELS | federated STG forecasting · Average over 5 clients, 60 min | MAE: 4.51 -> 4.21 | Improved |
| use SC-FSGL instead of FedAvg | federated STG forecasting · PEMSD4 | MAE: 3.41 -> 2.46 | Improved |
| use SC-FSGL instead of FUELS | federated STG forecasting · PEMSBAY | MAE: 2.34 -> 6.44 | Hurt |
| use SC-FSGL instead of Local training | federated STG forecasting · PEMSBAY | MAE: 2.39 -> 6.44 | Hurt |
| use SC-FSGL instead of Local training | federated STG forecasting · PEMSD7(M) | MAE: 7.11 -> 3.50 | Improved |
| use SC-FSGL instead of GMAN | federated STG forecasting · Average, 30 min horizon | MAE: 3.92 -> 3.67 | Improved |
| use SC-FSGL instead of GMAN | federated STG forecasting · Average, 15 min horizon | MAE: 3.64 -> 3.29 | Improved |
| use SC-FSGL instead of best baseline on average over clients | federated STG forecasting · Average, 60 min | MAPE: 10.18% -> 10.08% | Improved |
| remove causal codebook (ablation) | SC-FSGL · Average over 5 datasets | MAE | Hurt |
| remove conditional separation module (ablation) | SC-FSGL · Average over 5 datasets | MAE | Hurt |
Closest papers by meaning
- Less is More: Federated Graph Learning with Alleviating Topology Heterogeneity from A Causal Perspective (ICML 2025 · similarity 0.92)
- $S^2$FGL: Spatial Spectral Federated Graph Learning (ICML 2025 · similarity 0.89)
- Towards Effective Federated Graph Foundation Model via Mitigating Knowledge Entanglement (NeurIPS 2025 · similarity 0.89)
- Federated Graph Learning under Domain Shift with Generalizable Prototypes (AAAI 2024 · similarity 0.88)
- Federated Causality Learning with Explainable Adaptive Optimization (AAAI 2024 · similarity 0.88)
- FedSPA: Generalizable Federated Graph Learning under Homophily Heterogeneity (CVPR 2025 · similarity 0.88)
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.
