Causal structure learning on Synthetic nD=10000, d=2, dmax=10, 600 nodes
3.16Number of CI testsMB-by-MB+
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MB-by-MB+CI Test Type=d-separation2025.10 | 3.16 | 3.11 | 0 | 0.026 | |
| LDP+CI Test Type=d-separation2025.10 | 4.25 | 20.83 | 0 | 0.009 | |
| LOADCI Test Type=d-separation2025.10 | 4.36 | 4.15 | 100 | 0.003 | |
| LDECC+CI Test Type=d-separation2025.10 | 179.14 | 210.67 | 0 | 0.026 | |
| SNAP(∞)CI Test Type=d-separation2025.10 | 179.89 | 313.71 | 100 | 0.003 | |
| PCCI Test Type=d-separation2025.10 | 205.14 | 334.99 | 100 | 0.003 | |
| MB-by-MB+CI Test Type=G22025.10 | 3,200 | 16.22 | 0 | 0.031 | |
| LDP+CI Test Type=G22025.10 | 3,960 | 32.8 | 0 | 0.031 | |
| LOADCI Test Type=G22025.10 | 4,250 | 15.24 | 0.13 | 0.031 | |
| SNAP(∞)CI Test Type=G22025.10 | 179,700 | 130.66 | 0 | 0.054 | |
| PCCI Test Type=G22025.10 | 195,230 | 129.28 | 0.21 | 0.019 |