Research map / Probabilistic and scientific modeling / Differential equations and PDEs
A Deterministic Approximation to Neural SDEs
Andreas Look, Melih Kandemir, Barbara Rakitsch, Jan Peters
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 |
|---|---|---|---|
| use vertical moment matching instead of cubature (ablation) | NSDE-BMM · MNIST time series classification | accuracy: 95.16 -> 98.11 | Improved |
| use BMM instead of Monte Carlo sampling | NSDE · MNIST time series classification | accuracy: 81.60 -> 98.11 | Improved |
| use BMM instead of Monte Carlo sampling | NSDE · IMDB | accuracy: 84.92 -> 85.87 | Improved |
| add multi-step training instead of one-step training (ablation) | NSDE-BMM · Lotka-Volterra | MSE: 2.44 -> 1.75 | Improved |
| use BMM instead of cubature (ablation) | NSDE · 3-DOF-Robot (Transfer) | NLL: 352 -> 102 | Improved |
| use BMM instead of cubature (ablation) | NSDE · 3-DOF-Robot (IID) | NLL: 5.65 -> 7.27 | Hurt |
| use BMM instead of Monte Carlo sampling | NSDE · 3-DOF-Robot (Transfer) | NLL: 1821 -> 102 | Improved |
| increase integration time from 2s to 8s | NSDE-BMM · UCI Energy | NLL: 1.18 -> 0.70 | Improved |
| increase integration time from 2s to 8s | NSDE-MC · UCI Wine Red | NLL: 1.01 -> 1.06 | Hurt |
| use BMM instead of DVI | NSDE layer · UCI Energy | NLL: 1.01 -> 0.70 | Improved |
| use BMM instead of DVI | NSDE layer · UCI Naval | NLL: -6.29 -> -4.45 | Hurt |
| switch LSTM output from deterministic to stochastic | LSTM · Lotka-Volterra | MSE: 1.86 -> 2.41 | Hurt |
Closest papers by meaning
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions (NeurIPS 2025 · similarity 0.89)
- Efficient Training of Neural Stochastic Differential Equations by Matching Finite Dimensional Distributions (ICLR 2025 · similarity 0.88)
- Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data (ICLR 2024 · similarity 0.88)
- S$^{2}$-PINN: Stochastic Separable Physics-Informed Neural Networks (NeurIPS 2026 · similarity 0.87)
- Uncertainty Modeling in Graph Neural Networks via Stochastic Differential Equations (ICLR 2025 · similarity 0.86)
- On the Generalization and Approximation Capacities of Neural Controlled Differential Equations (ICLR 2024 · similarity 0.86)
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.
