Research map / Probabilistic and scientific modeling / Differential equations and PDEs
ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems
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 |
|---|---|---|---|
| Use ODIN instead of RKG3 | Lotka-Volterra parameter inference · Lotka-Volterra, σ = 0.1 | computation time (seconds): 79.0 ± 19.0 -> 13.4 ± 5.1 | Helped |
| Use ODIN instead of RKG3 | Lotka-Volterra parameter inference · Lotka-Volterra, σ = 0.5 | computation time (seconds): 76.5 ± 15.8 -> 11.4 ± 5.1 | Helped |
| Use ODIN instead of RKG3 | FitzHugh-Nagumo parameter inference · FitzHugh-Nagumo, SNR = 100 | computation time (seconds): 74.5 ± 14.3 -> 5.8 ± 3.5 | Helped |
| Use ODIN instead of RKG3 | FitzHugh-Nagumo parameter inference · FitzHugh-Nagumo, SNR = 10 | computation time (seconds): 77.5 ± 12.3 -> 4.4 ± 3.8 | Helped |
| Use ODIN instead of RKG3 | Protein Transduction parameter inference · Protein Transduction, σ = 0.001 | computation time (seconds): 469.0 ± 21.6 -> 8.9 ± 1.5 | Helped |
| Use ODIN instead of RKG3 | Protein Transduction parameter inference · Protein Transduction, σ = 0.01 | computation time (seconds): 480.0 ± 42.0 -> 20.6 ± 3.75 | Helped |
| Use ODE information as a constraint during regression | ODIN state inference · Benchmark systems, except high-noise Lotka-Volterra | state RMSE | Helped |
| Use ODE information as a constraint during regression | ODIN state inference · Lotka-Volterra, high noise | state RMSE | Hurt |
| Allow the derivative noise parameter γ to adapt during optimization | ODIN parameter inference · Protein Transduction | trajectory RMSE | Helped |
| Use γ as an optimized model-mismatch indicator | ODIN model selection · Lotka-Volterra candidate models | model mismatch detection: 10^{-6} ± 0.00; 3.01 ± 0.23; 10^{-6} ± 0.00; 3.03 ± 0.24 | Helped |
Closest papers by meaning
- ODEFormer: Symbolic Regression of Dynamical Systems with Transformers (ICLR 2024 · similarity 0.87)
- SING: SDE Inference via Natural Gradients (NeurIPS 2025 · similarity 0.87)
- Learning Physics Informed Neural ODEs with Partial Measurements (AAAI 2025 · similarity 0.86)
- Closing the Approximation Gap in Simulation-free Latent SDEs (NeurIPS 2026 · similarity 0.85)
- Exact Inference for Continuous-Time Gaussian Process Dynamics (AAAI 2024 · similarity 0.85)
- Generating Physical Dynamics under Priors (ICLR 2025 · similarity 0.85)
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.