Research map / Probabilistic and scientific modeling / Causal discovery
SCOUT: Cyclic Causal Discovery Under Soft Interventions with Unknown Targets
Alpar Turkoglu, Muralikrishnna Guruswamy Sethuraman, Faramarz Fekri
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 |
|---|---|---|---|
| add neural spline flow noise model (ablation) | SCOUT · Synthetic target recovery, Shift, Gaussian noise (d=10) | target recovery AUPRC: 0.144 -> 1.000 | Improved |
| add neural spline flow noise model (ablation) | SCOUT · Synthetic target recovery, Shift, Exponential noise (d=10) | target recovery AUPRC: 0.471 -> 1.000 | Improved |
| add neural spline flow noise model (ablation) | SCOUT · Synthetic target recovery, Scale, Gaussian noise (d=10) | target recovery AUPRC: 0.227 -> 1.000 | Improved |
| add neural spline flow noise model (ablation) | SCOUT · Synthetic target recovery, Noise intervention, Gaussian noise (d=10) | target recovery AUPRC: 0.393 -> 0.264 | Hurt |
| add neural spline flow noise model (ablation) | SCOUT · Synthetic target recovery, Noise intervention, Gumbel noise (d=10) | target recovery AUPRC: 0.463 -> 0.607 | Improved |
| increase graph size from d=10 to d=70 | BACKSHIFT · Shift target recovery, Gaussian noise | target recovery AUPRC: 0.90 -> 0.03 | Hurt |
| increase graph size from d=10 to d=70 | SCOUT · Shift target recovery, Gaussian noise | target recovery AUPRC: 1.00 -> 1.00 | No clear effect |
| use SCOUT instead of BACKSHIFT for noisy function interventions | SCOUT · Synthetic target recovery, Exponential noise (d=10) | target recovery AUPRC: 0.313 -> 0.590 | Improved |
| use SCOUT instead of BACKSHIFT on scale interventions with Exponential noise | SCOUT · Synthetic target recovery, Exponential noise (d=10) | target recovery AUPRC: 1.000 -> 0.975 | Hurt |
| use SCOUT instead of BACKSHIFT on SERGIO | SCOUT · SERGIO | target recovery AUPRC: 0.432 -> 0.978 | Improved |
Closest papers by meaning
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.90)
- Interventional Causal Discovery in a Mixture of DAGs (NeurIPS 2024 · similarity 0.90)
- Differentiable Cyclic Causal Discovery Under Unmeasured Confounders (NeurIPS 2025 · similarity 0.90)
- Learning Mixtures of Unknown Causal Interventions (NeurIPS 2024 · similarity 0.90)
- On Causal Discovery in the Presence of Deterministic Relations (NeurIPS 2024 · similarity 0.89)
- Adaptive Online Experimental Design for Causal Discovery (ICML 2024 · similarity 0.89)
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.
