Research map / Probabilistic and scientific modeling / Dynamical systems
Differentiable Sparse Identification of Lagrangian Dynamics
Zitong Zhang, Hao Sun
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 second-order derivative regularization on B-spline control points (ablation) | Ours · Single pendulum | l2 relative error x1e-2: 12.36 -> 0.41 | Improved |
| add second-order derivative regularization (ablation) | Ours · Spherical pendulum | l2 relative error x1e-2: 91.55 -> 2.00 | Improved |
| add second-order derivative regularization (ablation) | Ours · Double pendulum | l2 relative error x1e-2: 58.65 -> 1.15 | Improved |
| add second-order derivative regularization (ablation) | Ours · Magnetical pendulum | l2 relative error x1e-2: 40.55 -> 1.37 | Improved |
| randomly drop 5% of samples (ablation) | Ours · Single/Double/Spherical/Chaos pendulum | l2 relative error x1e-2: 0.41 -> 0.41 (single pendulum) | No clear effect |
| drop 5% of samples on magnetical pendulum (ablation) | Ours · Magnetical pendulum | l2 relative error x1e-2: 1.37 -> 3.01 | Hurt |
| use B-spline surrogate with physics loss instead of xL-SINDy under noise | Lagrangian identification · Double pendulum at 10% noise | l2 relative error x1e-2: 73.42 -> 21.67 | Improved |
| use B-spline method instead of xL-SINDy on noise-free data | Single pendulum · Single pendulum, 0% noise | l2 relative error x1e-2: 0.31 -> 0.41 | No clear effect |
| use B-spline method instead of xL-SINDy | Sphere spring pendulum · avg over noise levels | l2 relative error x1e-2: 2.11 -> 2.03 | No clear effect |
| use B-spline method instead of uDSR symbolic regression | Single pendulum · avg over noise levels | l2 relative error x1e-2: 160.7 -> 4.22 | Improved |
Closest papers by meaning
- Learned Lagrangian Models of PDEs via Euler–Lagrange Residual Minimization (NeurIPS 2026 · similarity 0.85)
- Symmetry-Informed Governing Equation Discovery (NeurIPS 2024 · similarity 0.84)
- Identification of Analytic Nonlinear Dynamical Systems with Non-asymptotic Guarantees (NeurIPS 2024 · similarity 0.84)
- Extracting Governing Equations from Latent Dynamics via Multi-View Contrastive Learning (NeurIPS 2026 · similarity 0.84)
- A Riemannian Framework for Learning Reduced-order Lagrangian Dynamics (ICLR 2025 · similarity 0.84)
- Stable Port-Hamiltonian Neural Networks (NeurIPS 2025 · similarity 0.83)
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.
