Research map / Probabilistic and scientific modeling / Bayesian optimization
Bayesian Network Learning via Topological Order
Young Woong Park, Diego Klabjan
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 |
|---|---|---|---|
| use topological-order search (GD10) instead of arc-based search (DIST) on dense solutions | Gaussian Bayesian network learning · Sachs flow cytometry (lambda=0.05) | delta_sol: 1.87% -> 0.32% | Improved |
| use GD10 instead of DIST on real data at lambda 0.25 | Gaussian Bayesian network learning · Sachs flow cytometry (lambda=0.25) | delta_sol: 0.94% -> 0.00% | Improved |
| use topological-order algorithms instead of DIST for dense solutions | GD10 and IR10 · synthetic data, density ln<2 vs denser | delta_sol | Improved |
| use DIST instead of topological-order algorithms for sparse solutions | DIST · synthetic data, ln density below 2 | average delta_sol: 0.06%, 0.15%, and 0.04% | No clear effect |
| use DIST instead of GD10/IR10 on very sparse high-dimensional data | DIST · high dimensional synthetic | time | Improved |
| lower lambda (more arcs) favors topological-order algorithms | GD10 and IR10 · sparse synthetic data | delta_sol | Improved |
| use topological-order algorithms instead of DIST regarding arcs selected | GD10 and IR10 · dense synthetic data | number of selected arcs | Improved |
| use MIPin instead of MIPto/MIPcp at small m | MIPin · sparse synthetic, m=20 and 30 | delta_sol | Improved |
| increase m and s and decrease lambda for MIPin | MIPin · sparse synthetic, n=100 | delta_sol | Hurt |
| use MIPcp instead of MIPto at m=40 or s=3 | MIPcp · sparse synthetic, n=100 | delta_sol | Improved |
| use iterative algorithms instead of MIP models | GD10 and IR10 · sparse synthetic, n=100, m in 20-40 | delta_sol and time | Improved |
| use GD10 instead of IR10 | topological-order algorithms · synthetic data | delta_sol | No clear effect |
Closest papers by meaning
- Efficient Bayesian Network Structure Learning via Parameterized Local Search on Topological Orderings (AAAI 2021 · similarity 0.90)
- Consistent Second-Order Conic Integer Programming for Learning Bayesian Networks (JMLR 2023 · similarity 0.88)
- Learning Gaussian DAGs from Network Data (JMLR 2024 · similarity 0.86)
- Optimizing Regularized Cholesky Score for Order-Based Learning of Bayesian Networks (TPAMI 2020 · similarity 0.85)
- The Complexity of Bayesian Network Learning: Revisiting the Superstructure (NeurIPS 2021 · similarity 0.85)
- Bayesian Optimization of Functions over Node Subsets in Graphs (NeurIPS 2024 · similarity 0.84)
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.
