Research map / Graph learning / Knowledge and temporal graphs
Graph Neural Controlled Differential Equations for Traffic Forecasting
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| combine temporal and spatial NCDE processing instead of using either alone (ablation) | STG-NCDE · PeMSD3, PeMSD4, PeMSD7, PeMSD8, PeMSD7(M), PeMSD7(L) | MAE: Only temporal: 20.44 vs 15.57; Only spatial: 15.92 vs 15.57 (PeMSD3) | Helped |
| add spatial processing (ablation) | STG-NCDE temporal-only variant · PeMSD3 | MAE: 20.44 -> 15.57 | Helped |
| add temporal processing (ablation) | STG-NCDE spatial-only variant · PeMSD3 | MAE: 15.92 -> 15.57 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD3 | MAE: 16.64 -> 15.57 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD4 | MAE: 19.50 -> 19.21 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD7 | MAE: 21.77 -> 20.53 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD8 | MAE: 15.76 -> 15.45 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD7(M) | MAE: 2.75 -> 2.68 | Helped |
| use STG-NCDE instead of Z-GCNETs | traffic forecasting · PeMSD7(L) | MAE: 2.91 -> 2.87 | Helped |
| increase node-embedding size C up to 10 (ablation) | STG-NCDE · PeMSD7 | MAE | Helped |
| increase missing sensing values from 10% to 50% | STG-NCDE · irregular PeMSD4 | MAE: 19.36 -> 19.98 | Hurt |
| increase missing sensing values from 10% to 50% | STG-NCDE · irregular PeMSD8 | MAE: 15.68 -> 16.68 | Hurt |
Closest papers by meaning
- Learnable Path in Neural Controlled Differential Equations (AAAI 2023 · similarity 0.90)
- CSG-ODE: ControlSynth Graph ODE For Modeling Complex Evolution of Dynamic Graphs (ICML 2025 · similarity 0.87)
- Efficient Traffic Prediction Through Spatio-Temporal Distillation (AAAI 2025 · similarity 0.87)
- Designing Specialized Two-Dimensional Graph Spectral Filters for Spatial-Temporal Graph Modeling (AAAI 2025 · similarity 0.87)
- Signed Graph Neural Ordinary Differential Equation for Modeling Continuous-Time Dynamics (AAAI 2024 · similarity 0.86)
- Spatio-Temporal Pivotal Graph Neural Networks for Traffic Flow Forecasting (AAAI 2024 · similarity 0.86)
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.