Research map / Probabilistic and scientific modeling / Causal discovery
Minimal I-MAP MCMC for Scalable Structure Discovery in Causal DAG Models
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 minimal I-MAP MCMC instead of order MCMC | Bayesian DAG structure recovery · Synthetic n=1000, p=30 | AUROC, compelled features: .395 -> .958 | Helped |
| Use minimal I-MAP MCMC instead of partition MCMC | Bayesian DAG structure recovery · Synthetic n=1000, p=30 | AUROC, undirected features: .890 -> 1.00 | Helped |
| Use minimal I-MAP MCMC instead of MMHC bootstrap | Bayesian DAG structure recovery · Dream4 multifactorial | AUROC, undirected features: .552 -> .574 | Hurt |
| Add a path prior to the order prior (ablation) | minIMAP · Protein network | AUROC, directed edge recovery: .917 -> .929 | Helped |
| Apply partial bias correction rather than no bias correction | Order MCMC · Synthetic n=100, p=30 | AUROC, undirected features: .957 -> .956 | No clear effect |
| Apply full bias correction rather than no bias correction | Order MCMC · Dream4 multifactorial | AUROC, undirected features: .599 -> .563 | Hurt |
| Use minimal I-MAP MCMC rather than Order MCMC | MCMC estimation of directed feature probabilities · Synthetic datasets, n=1000, p=30 | Average correlation between seeded runs: .928 -> .977 | Helped |
Closest papers by meaning
- BCD Nets: Scalable Variational Approaches for Bayesian Causal Discovery (NeurIPS 2021 · similarity 0.92)
- Causal Discovery via Bayesian Optimization (ICLR 2025 · similarity 0.91)
- ProDAG: Projected Variational Inference for Directed Acyclic Graphs (NeurIPS 2025 · similarity 0.89)
- Boosting Causal Structure Learning: An Asymmetric Exponential Modulation Gaussian-Based Adaptive Sample Reweighting Framework (AAAI 2025 · similarity 0.89)
- Sample Efficient Bayesian Learning of Causal Graphs from Interventions (NeurIPS 2024 · similarity 0.89)
- A Meta-Learning Approach to Bayesian Causal Discovery (ICLR 2025 · 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.