Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Gaussian processes at the Helm(holtz): A more fluid model for ocean currents
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 |
|---|---|---|---|
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Vortex | RMSE (velocity F): 0.72 -> 0.24 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Vortex | RMSE (divergence δ): 0.22 -> 0.0 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Small divergence | RMSE (velocity F): 1.25 -> 1.11 | Mixed |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Small divergence | RMSE (divergence δ): 1.45 -> 2.62 | Hurt |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Medium divergence | RMSE (velocity F): 0.19 -> 0.17 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Medium divergence | RMSE (divergence δ): 0.33 -> 0.39 | Hurt |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Big divergence | RMSE (velocity F): 0.10 -> 0.04 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Big divergence | RMSE (divergence δ): 0.12 -> 0.05 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Duffing with small divergence | RMSE (velocity F): 2.05 -> 0.96 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Duffing with small divergence | RMSE (divergence δ): 0.95 -> 0.94 | No clear effect |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Duffing with medium divergence | RMSE (divergence δ): 0.50 -> 0.14 | Helped |
| Model the Helmholtz potentials with separate GPs instead of modeling velocity components independently | SE-Helmholtz GP versus SE-velocity GP · Duffing with big divergence | RMSE (velocity F): 0.22 -> 0.41 | Hurt |
Closest papers by meaning
- OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting (NeurIPS 2026 · similarity 0.85)
- HHD-GP: Incorporating Helmholtz-Hodge Decomposition into Gaussian Processes for Learning Dynamical Systems (NeurIPS 2024 · similarity 0.84)
- Physics-Informed Variational State-Space Gaussian Processes (NeurIPS 2024 · similarity 0.83)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · similarity 0.83)
- Geometric and Physical Constraints Synergistically Enhance Neural PDE Surrogates (ICML 2025 · similarity 0.83)
- Implicit Gaussian process representation of vector fields over arbitrary latent manifolds (ICLR 2024 · similarity 0.82)
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.