Research map / Probabilistic and scientific modeling / Causal discovery
IDYNO: Learning Nonparametric DAGs from Interventional Dynamic Data
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 |
|---|---|---|---|
| add interventional loss (separate distribution for intervened nodes) to linear DYNOTEARS-style objective | IDYNO (linear) · Synthetic linear interventional | SHD: 32 -> 19 | Helped |
| linear IDYNO vs DYNOTEARS on purely observational data | IDYNO (linear) · Synthetic linear observational | SHD: 2.0 vs 2.0 | No clear effect |
| switch from linear to neural MLP model (IDYNO-nn) on hard-intervention nonlinear data | IDYNO · Nonlinear hard-intervention synthetic (d=5 to 20) | SHD | Helped |
| soft-intervention loss function added to IDYNO-nn (ablation) | IDYNO-nn · Nonlinear soft-intervention synthetic (d=20) | SHD: 39.0 (IDYNO-nn-soft) vs IDYNO-nn (much worse) | Helped |
| neural model with hard-intervention loss on network administration ring DBN | IDYNO-nn · Network administration (ring topology) | SHD | Helped |
| neural soft-intervention variant IDYNO-nn-soft vs IDYNO-nn on Lunar Lander (ablation) | IDYNO-nn · Continuous Lunar Lander | SHD: 31.2 -> 27.4 | Helped |
| interventional data use in linear IDYNO vs DYNOTEARS on Lunar Lander | IDYNO (linear) · Continuous Lunar Lander | SHD: 32.6 -> 32.2 | Helped |
Closest papers by meaning
- Meta-D2AG: Causal Graph Learning with Interventional Dynamic Data (NeurIPS 2025 · similarity 0.90)
- Causal Discovery via Bayesian Optimization (ICLR 2025 · similarity 0.90)
- Effective Causal Discovery under Identifiable Heteroscedastic Noise Model (AAAI 2024 · similarity 0.89)
- Interventional Causal Discovery in a Mixture of DAGs (NeurIPS 2024 · similarity 0.88)
- Sample Efficient Bayesian Learning of Causal Graphs from Interventions (NeurIPS 2024 · similarity 0.88)
- Coarsening Linear Non-Gaussian Causal Models with Cycles (NeurIPS 2026 · similarity 0.88)
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.