Research map / Probabilistic and scientific modeling / Causal discovery
Identifiability and Asymptotics in Learning Homogeneous Linear ODE Systems from Discrete Observations
Yuanyuan Wang, Wei Huang, Mingming Gong, Xi Geng, Tongliang Liu, Kun Zhang, Dacheng Tao
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 |
|---|---|---|---|
| aggregate observations with k=20 instead of using original observations | NLS estimator · Simulated 3-D ODE, n=100 | MSE: 0.480 -> 0.858 | Hurt |
| aggregate observations with k=20 instead of using original observations | NLS estimator · Simulated 3-D ODE, n=2000 | MSE: 0.023 -> 0.023 | No clear effect |
| aggregate observations with k=20 instead of using original observations | NLS estimator · Simulated 3-D ODE, n=100 | Confidence region coverage %: 94 -> 30.5 | Hurt |
| aggregate observations with k=5 instead of original | NLS estimator · Simulated 3-D ODE, n=100 | Type II error rate % for a33: 82.5 -> 88.5 | Hurt |
| aggregate observations with k=20 instead of original | NLS estimator · Simulated 3-D ODE, n=100 | Type II error rate % for a33: 82.5 -> 99 | Hurt |
| increase sample size from 100 to 2000 (original observations) | NLS estimator · Simulated 3-D ODE | MSE: 0.480 -> 0.023 | Improved |
| increase sample size from 100 to 2000 (original observations) | NLS estimator · Simulated 3-D ODE | Type II error rate % for a33: 82.5 -> 0 | Improved |
| rescale time by k (0.01 to 100) instead of original time scale | NLS estimator · Simulated 3-D ODE, n=100 | MSE: 0.480 -> 0.419 | No clear effect |
| increase aggregation from k=5 to k=20 | NLS estimator · Simulated 3-D ODE, n=1000 | MSE: 0.045 -> 0.045 | No clear effect |
Closest papers by meaning
- Identifiability Analysis of Linear ODE Systems with Hidden Confounders (NeurIPS 2024 · similarity 0.91)
- Disentangling Continuous-Time Latent Dynamics: Identifiability of Latent SDEs via Diffusion Shifts (NeurIPS 2026 · similarity 0.85)
- ODEFormer: Symbolic Regression of Dynamical Systems with Transformers (ICLR 2024 · similarity 0.85)
- On the Identifiability of Switching Dynamical Systems (ICML 2024 · similarity 0.85)
- Zero-shot Imputation with Foundation Inference Models for Dynamical Systems (ICLR 2025 · similarity 0.85)
- Global Convergence in Neural ODEs: Impact of Activation Functions (ICLR 2025 · 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.
