Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Learning Bayesian Networks from Ordinal Data
Xiang Ge Luo, Giusi Moffa, Jack Kuipers
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 |
|---|---|---|---|
| model ordinal data with a latent Gaussian structure via OSEM instead of PC or BDe/BGe methods | structure learning from ordinal data · Simulated DAGs, ROC of pattern recovery | TPR vs FPRp | Improved |
| use BGe score instead of BDe score for ordinal data | hybrid structure learning · Simulated DAGs, ROC of pattern recovery | TPR vs FPRp | Improved |
| increase expected number of ordinal levels | OSEM and BGe · 20 nodes, N=500 | ROC performance | Improved |
| increase expected number of ordinal levels | hybrid BDe · 20 nodes, N=500 | ROC performance | Hurt |
| increase Dirichlet concentration nu from 0.8 to 2 (more even thresholds) | all hybrid methods · 20 nodes, N=500 | ROC performance | Improved |
| denser networks, 2 to 5 expected neighbours | all methods · 20 nodes, N=500 | average TPR: almost 75% at d=2 -> below 75% at d=5 | Hurt |
| use OSEM instead of hybrid BDe/BGe | runtime · 12, 20, 30 nodes, N=500 | runtime: less than 10 seconds -> around 1.5, 2.5 and 3.5 minutes | Hurt |
| OSEM versus BDe baseline on real data | predictive log loss · five real ordinal datasets, 80/20 split | relative log loss | Improved |
| use ordinal OPCART instead of PCART | mixed-data structure learning · simulated ordinal data | structure recovery | Hurt |
Closest papers by meaning
- Exploiting Experts’ Knowledge for Structure Learning of Bayesian Networks (TPAMI 2016 · similarity 0.86)
- Scores for Learning Discrete Causal Graphs with Unobserved Confounders (AAAI 2024 · similarity 0.84)
- Optimizing Regularized Cholesky Score for Order-Based Learning of Bayesian Networks (TPAMI 2020 · similarity 0.84)
- Learning the Parameters of Bayesian Networks from Uncertain Data (AAAI 2021 · similarity 0.84)
- Bayesian Causal Structural Learning with Zero-Inflated Poisson Bayesian Networks (NeurIPS 2020 · similarity 0.84)
- Local Causal Structure Learning in the Presence of Latent Variables (ICML 2024 · similarity 0.83)
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.
