Research map / Time series / Causal and generative time series
Neural ODE Processes
Alexander Norcliffe, Cristian Bodnar, Ben Day, Jacob Moss, Pietro Liò
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 NDP instead of NP | 1D regression · Sine | MSE × 10^-2: 5.93 $\pm$ 0.96 -> 2.09 $\pm$ 0.12 | Improved |
| use NDP instead of NP | 1D regression · Linear | MSE × 10^-2: 5.85 $\pm$ 0.70 -> 3.76 $\pm$ 0.32 | Improved |
| use NDP instead of NP | 1D regression · Exponential | MSE × 10^-2: 0.29 $\pm$ 0.03 -> 0.31 $\pm$ 0.08 | No clear effect |
| use NDP instead of NP | 1D regression · Oscillators | MSE × 10^-2: 0.64 $\pm$ 0.06 -> 0.72 $\pm$ 0.08 | No clear effect |
| use NDP instead of NP | Lotka-Volterra · Lotka-Volterra | test MSE × 10^-2: 44 \pm 4 -> 15 \pm 2 | Improved |
| use NDP instead of NP | CharacterTrajectories · CharacterTrajectories | test MSE × 10^-1: 4.6 \pm 0.1 -> 3.4 \pm 0.1 | Improved |
| use NDP instead of NP on variable rotating digits | Variable Rotating MNIST · Variable Rotating MNIST | qualitative prediction quality | Improved |
| use NDP instead of NP on constant-velocity rotating digits | Rotating MNIST · Rotating MNIST | qualitative prediction quality | No clear effect |
| use second-order latent dynamics (ablation) | NDP · Sine | MSE × 10^-2: 2.09 $\pm$ 0.12 -> 2.75 $\pm$ 0.19 | Hurt |
| use second-order latent dynamics (ablation) | NDP · Linear | MSE × 10^-2: 3.76 $\pm$ 0.32 -> 4.37 $\pm$ 1.14 | Hurt |
| use a linear latent-only decoder (ablation) | NDP · Sine | MSE × 10^-2: 2.09 $\pm$ 0.12 -> 2.51 $\pm$ 0.24 | Hurt |
| set latent ODE dimension to 1 instead of 2 (ablation) | NDP on Sine · Sine | MSE × 10^-2: 1.7 $\pm$ 0.1 -> 5.6 $\pm$ 1.3 | Hurt |
Closest papers by meaning
- Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data (ICLR 2024 · similarity 0.91)
- Flow Matching Neural Processes (NeurIPS 2025 · similarity 0.89)
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions (NeurIPS 2025 · similarity 0.88)
- ControlSynth Neural ODEs: Modeling Dynamical Systems with Guaranteed Convergence (NeurIPS 2024 · similarity 0.88)
- How Deep Do We Need: Accelerating Training and Inference of Neural ODEs via Control Perspective (ICML 2024 · similarity 0.88)
- KoNODE: Koopman-Driven Neural Ordinary Differential Equations with Evolving Parameters for Time Series Analysis (ICML 2025 · 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.
