Research map / Probabilistic and scientific modeling / Molecules and proteins
Riemannian Neural SDE: Learning Stochastic Representations on Manifolds
Sung Woo Park, Hyomin Kim, Kyungjae Lee, Junseok Kwon
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 |
|---|---|---|---|
| replace deterministic RNODE with stochastic RNSDE (ablation) | RNSDE · Sphere 8-shapes | W2 (x10^-2): 8.01 -> 5.67 | Improved |
| replace deterministic RNODE with stochastic RNSDE (ablation) | RNSDE · Sphere two moons | W2 (x10^-2): 7.68 -> 5.66 | Improved |
| replace deterministic RNODE with stochastic RNSDE (ablation) | RNSDE · Sphere spiral | W2 (x10^-2): 9.32 -> 6.97 | Improved |
| replace deterministic RNODE with stochastic RNSDE (ablation) | RNSDE · Torus helix coil | W2 (x10^-2): 3.18 -> 2.86 | Improved |
| RNSDE instead of RCPM | RNSDE · Torus helix coil | W2 (x10^-2): 7.64 -> 2.86 | Improved |
| RNSDE instead of Moser flow | RNSDE · Sphere spiral | W2 (x10^-2): 7.56 -> 6.97 | Improved |
| RNSDE instead of MCNF | RNSDE · Sphere 8-shapes | W2 (x10^-2): 11.26 -> 5.67 | Improved |
| RNSDE instead of RNODE for shape interpolation | RNSDE · Shape interpolation (i)->(ii) | W2 (x10^-2): 7.09 -> 6.09 | Improved |
| RNSDE instead of RNODE for shape interpolation | RNSDE · Shape interpolation (i)->(iii) | W2 (x10^-2): 9.01 -> 7.61 | Improved |
| RNSDE instead of RCPM on volcano data | RNSDE · Volcano eruption | W2 (x10^-1): 1.83 -> 1.08 | Improved |
| RNSDE instead of Latent ODE on vessel routes | RNSDE · Vessel route | mean geodesic error: 14.25 -> 12.83 | Improved |
| RNSDE instead of ODE-RNN on vessel routes | RNSDE · Vessel route | W2: 7.29 -> 6.23 | Improved |
Closest papers by meaning
- Efficient Training of Neural Stochastic Differential Equations by Matching Finite Dimensional Distributions (ICLR 2025 · similarity 0.85)
- Data-free Neural Representation Compression with Riemannian Neural Dynamics (ICML 2024 · similarity 0.85)
- Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data (ICLR 2024 · similarity 0.85)
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions (NeurIPS 2025 · similarity 0.84)
- Generative Modeling on Manifolds Through Mixture of Riemannian Diffusion Processes (ICML 2024 · similarity 0.84)
- Building Transformation Layers for Riemannian Neural Networks (NeurIPS 2026 · similarity 0.84)
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.
