Research map / Probabilistic and scientific modeling / Dynamical systems
Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamics
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 |
|---|---|---|---|
| Use NCLaw instead of GNN | PDE dynamics learned from one motion trajectory · Generalization (Overall) | MSE: 1.2e-1±2.6e-1 -> 1.9e-4±2.3e-4 | Helped |
| Use NCLaw instead of Neural | PDE dynamics learned from one motion trajectory · Generalization (Overall) | MSE: 4.9e-2±8.8e-2 -> 1.9e-4±2.3e-4 | Helped |
| Use NCLaw instead of GNN | Material dynamics reconstruction · Reconstruction (Overall) | MSE: 1.8e-2±1.0e-2 -> 8.6e-5±1.0e-4 | Helped |
| Add rotation equivariance to NCLaw (ablation) | Neural constitutive laws · Plasticine | training convergence | Helped |
| Add undeformed state equilibrium to NCLaw (ablation) | Neural constitutive laws · Plasticine | simulation stability | Helped |
| Use NCLaw instead of the labeled oracle | Water parameter identification · Water | stress P MSE: 2.6e-3 -> 6.1e-1 | Hurt |
| Use NCLaw instead of the labeled oracle | Water parameter identification · Water | F_new MSE: 2.381e-6 -> 2.388e-6 | No clear effect |
| Use Neural instead of NCLaw | Material dynamics reconstruction · Jell-O | MSE: 2.4e-4 -> 1.2e-5 | Helped |
Closest papers by meaning
- PARCv2: Physics-aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics Modeling (ICML 2024 · similarity 0.88)
- UniPhy: Learning a Unified Constitutive Model for Inverse Physics Simulation (CVPR 2025 · similarity 0.88)
- Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systems (CVPR 2025 · similarity 0.87)
- Mechanistic PDE Networks for Discovery of Governing Equations (ICML 2025 · similarity 0.87)
- Riemann Tensor Neural Networks: Learning Conservative Systems with Physics-Constrained Networks (ICML 2025 · similarity 0.87)
- Harnessing the Power of Neural Operators with Automatically Encoded Conservation Laws (ICML 2024 · 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.