Research map / Graph learning / Knowledge and temporal graphs
Enabling arbitrary inference in spatio-temporal dynamic systems: A physics-inspired perspective
Yan Ge, Zhengyang Zhou, Qihe Huang, Yuxuan Liang, Yang Wang
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 Time-Gated Spectral Segmentation Perception (ablation) | PhySTA · PEMS-BAY (mask=0 and 0.5) | MAE: 1.80 -> 1.66 and 4.12 -> 3.52 | Improved |
| add GT-FNO (ablation) | PhySTA · PEMS-BAY (mask=0 and 0.5) | MAE: 2.05 -> 1.66 and 5.03 -> 3.52 | Improved |
| add edge-node coupled convolution (ENCC) (ablation) | PhySTA · PEMS-BAY (mask=0 and 0.5) | MAE: 1.73 -> 1.66 and 3.87 -> 3.52 | Improved |
| add multi-scale graph convolution (MSGCN) (ablation) | PhySTA · PEMS-BAY (mask=0.5) | MAE: 3.96 -> 3.52 | Improved |
| add Adaptive Multi-scale Interaction (AMI) module (ablation) | PhySTA · PEMS-BAY (mask=0 and 0.5) | MAE: 1.88 -> 1.66 and 4.13 -> 3.52 | Improved |
| add AMI module (ablation) | PhySTA · PEMS-BAY (mask=0.5) | RMSE: 8.26 -> 7.04 | Improved |
| compare PhySTA with STTN under 30% masking | PhySTA vs STTN · SD (mask=0.3) | MAE: 55.73 -> 50.34 | Improved |
| compare PhySTA with STTN under 70% masking | PhySTA vs STTN · SD (mask=0.7) | MAE: 108.46 -> 96.09 | Improved |
| compare PhySTA with AGCRN without masking | PhySTA vs AGCRN · PEMS-BAY (mask=0) | MAE: 1.64 vs 1.66 | No clear effect |
| increase global edges per node (topk edge) (ablation) | PhySTA · PEMS-BAY and KnowAir (high masking) | MAE | Hurt |
| stack more edge-node coupling layers (ecc layer) (ablation) | PhySTA · KnowAir | MAE | Hurt |
| increase GT-FNO hidden dimension (ablation) | PhySTA · PEMS-BAY | MAE | Improved |
Closest papers by meaning
- Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems (NeurIPS 2026 · similarity 0.88)
- Structural Inference of Dynamical Systems with Conjoined State Space Models (NeurIPS 2024 · similarity 0.86)
- Breaking the Discretization Barrier of Continuous Physics Simulation Learning (NeurIPS 2025 · similarity 0.86)
- Learning Spatiotemporal Dynamical Systems from Point Process Observations (ICLR 2025 · similarity 0.86)
- Benchmarking Structural Inference Methods for Interacting Dynamical Systems with Synthetic Data (NeurIPS 2024 · similarity 0.86)
- A Graph Dynamics Prior for Relational Inference (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.
