Research map / Probabilistic and scientific modeling / Differential equations and PDEs
DGNet: Learning Spatiotemporal PDEs with Discrete Green Networks
Yingjie Tan, Quanming Yao, Yaqing 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 discrete Green solver (vs generic end-to-end GNN) (ablation) | DGNet · Complex Obstacles | MSE: 4.23e-02 -> 6.69e-05 | Improved |
| add discrete Green solver (vs generic end-to-end GNN) (ablation) | DGNet · FitzHugh-Nagumo | MSE: 5.39e-05 -> 1.18e-07 | Improved |
| add discrete Green solver (vs generic end-to-end GNN) (ablation) | DGNet · Laser Heat | MSE: 6.12e+03 -> 1.76e+01 | Improved |
| add physics prior operator L_physics (ablation) | DGNet · Complex Obstacles | MSE: 8.12e-03 -> 6.69e-05 | Improved |
| add neural correction operator L_neural (ablation) | DGNet · Complex Obstacles | MSE: 6.10e-04 -> 6.69e-05 | Improved |
| add residual GNN (ablation) | DGNet · Complex Obstacles | MSE: 1.43e-04 -> 6.69e-05 | Improved |
| add physics prior operator L_physics (ablation) | DGNet · FitzHugh-Nagumo | MSE: 7.24e-06 -> 1.18e-07 | Improved |
| add physics prior operator L_physics (ablation) | DGNet · Laser Heat | MSE: 1.32e+03 -> 1.76e+01 | Improved |
| add neural correction operator L_neural (ablation) | DGNet · Laser Heat | MSE: 5.93e+02 -> 1.76e+01 | Improved |
| add residual GNN (ablation) | DGNet · FitzHugh-Nagumo | MSE: 4.64e-07 -> 1.18e-07 | Improved |
| use DGNet instead of best baseline | neural PDE solvers · Allen-Cahn | MSE: 2.60e-01 -> 8.75e-03 | Improved |
| use DGNet instead of best baseline on unseen laser source terms | neural PDE solvers · Laser Heat | MSE: 1.95e+03 -> 1.76e+01 | Improved |
Closest papers by meaning
- PhyMPGN: Physics-encoded Message Passing Graph Network for spatiotemporal PDE systems (ICLR 2025 · similarity 0.92)
- P$^2$C$^2$Net: PDE-Preserved Coarse Correction Network for efficient prediction of spatiotemporal dynamics (NeurIPS 2024 · similarity 0.91)
- Mechanistic PDE Networks for Discovery of Governing Equations (ICML 2025 · similarity 0.88)
- MultiPDENet: PDE-embedded Learning with Multi-time-stepping for Accelerated Flow Simulation (ICML 2025 · similarity 0.88)
- Text2PDE: Latent Diffusion Models for Accessible Physics Simulation (ICLR 2025 · similarity 0.88)
- CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs (NeurIPS 2025 · similarity 0.87)
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.
