Research map / Time series / Long-term forecasting
Adaptive Graph Convolutional Recurrent Network 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 |
|---|---|---|---|
| replace shared GCN parameters with node-adaptive parameters (NAPL) (ablation) | GCGRU · PeMSD4, long-term horizons | prediction performance | Helped |
| replace shared GCN parameters with node-adaptive parameters (NAPL) (ablation) | GCGRU · PeMSD4, short-term horizons | prediction performance | Hurt |
| infer the graph from data with DAGG instead of using a pre-defined graph (ablation) | GCGRU · PeMSD4 | prediction performance | Helped |
| share one node embedding across NAPL-GCN layers and DAGG (ablation) | AGCRN-I · PeMSD4 | prediction performance | Helped |
| remove the identity matrix from DAGG (ablation) | AGCRN-I · PeMSD4 (Average) | MAE: 20.18 -> 23.35 | Hurt |
| remove the identity matrix from DAGG (ablation) | AGCRN-I · PeMSD4 (Average) | RMSE: 32.30 -> 37.07 | Hurt |
| remove the identity matrix from DAGG (ablation) | AGCRN-I · PeMSD4 (Average) | MAPE: 13.70% -> 15.82% | Hurt |
| use second-order Chebyshev expansion in DAGG (ablation) | AGCRN-I · PeMSD4 (Average) | MAE: 20.18 -> 20.11 | No clear effect |
| vary the node-embedding dimension (ablation) | AGCRN · PeMSD4 | prediction performance | Mixed |
Closest papers by meaning
- Dynamic Spatial-Temporal Graph Convolutional Neural Networks for Traffic Forecasting (AAAI 2019 · similarity 0.93)
- Gated Residual Recurrent Graph Neural Networks for Traffic Prediction (AAAI 2019 · similarity 0.92)
- Efficient Traffic Prediction Through Spatio-Temporal Distillation (AAAI 2025 · similarity 0.90)
- Spatio-Temporal Pivotal Graph Neural Networks for Traffic Flow Forecasting (AAAI 2024 · similarity 0.89)
- TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts (ICLR 2024 · similarity 0.89)
- Towards Dynamic Spatial-Temporal Graph Learning: A Decoupled Perspective (AAAI 2024 · 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.