Research map / Probabilistic and scientific modeling / Causal discovery
Mitigating Prior Errors in Causal Structure Learning: A Resilient Approach via Bayesian Networks
Lyuzhou Chen, Taiyu Ban, Xiangyu Wang, Derui Lyu, Huanhuan Chen
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 quasi-circle-based prior correction | HC · CHILD (0 right, 4 wrong priors) | SHD: 15.72 -> 11.85 | Improved |
| add quasi-circle-based prior correction | HC · ALARM (0 right, 4 wrong priors) | SHD: 30.88 -> 22.93 | Improved |
| add quasi-circle-based prior correction | HC · INSURANCE (0 right, 4 wrong priors) | SHD: 37.52 -> 32.40 | Improved |
| add quasi-circle-based prior correction when all priors are correct | HC · CHILD (4 right, 0 wrong) | SHD: 8.55 -> 8.90 | Hurt |
| add quasi-circle-based prior correction when all priors are correct | HC · ALARM (2 right, 0 wrong) | SHD: 20.87 -> 21.33 | Hurt |
| add quasi-circle correction to NOTEARS | NOTEARS · Synthetic 30 nodes ER=1 DR=10 | F1: 91.6 -> 100.0 | Improved |
| add quasi-circle correction to NOTEARS with error ratio 1 | NOTEARS · Synthetic 30 nodes ER=1 DR=10, 30% correct priors | F1: 76.0 -> 80.4 | Improved |
| add quasi-circle correction to NOTEARS under many prior errors | NOTEARS · Synthetic 40 nodes ER=2 DR=20, 30% priors, error ratio 1 | SHD: 50.3 -> 14.8 | Improved |
| add quasi-circle correction to NOTEARS | NOTEARS · SACHS real, 20% priors, error ratio 0.5 | SHD: 27 -> 20 | Improved |
| add quasi-circle correction to NOTEARS | NOTEARS · MAGIC-NIAB real, 20% priors, error ratio 0.5 | F1: 27.8 -> 19.6 | Hurt |
| add quasi-circle correction to HC (prior-quality lower bound) | HC · CHILD, 4 prior errors | required prior quality: 58.3 -> 47.7 | Improved |
| add quasi-circle correction to HC (prior-quality lower bound), 1 error | HC · ALARM, 1 prior error | required prior quality: 80.2 -> 85.5 | Hurt |
Closest papers by meaning
- A Meta-Learning Approach to Bayesian Causal Discovery (ICLR 2025 · similarity 0.89)
- Local Causal Structure Learning in the Presence of Latent Variables (ICML 2024 · similarity 0.88)
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.88)
- Estimating Interventional Distributions with Uncertain Causal Graphs through Meta-Learning (NeurIPS 2025 · similarity 0.88)
- Standardizing Structural Causal Models (ICLR 2025 · similarity 0.88)
- Learning Structural Causal Models from Ordering: Identifiable Flow Models (AAAI 2025 · 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.
