Node-level Regression Uncertainty Quantification on Grid
100PICPER-GNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ER-GNN2026.05 | 100 | 11.92 | |
| BayesianNN2026.05 | 100 | 3.01 | |
| QpiGNNWidth penalty (lambda)=opt.2026.05 | 99 | 0.93 | |
| QpiGNNWidth penalty (lambda)=0.52026.05 | 98 | 0.87 | |
| QpiGNNWidth penalty (lambda)=0.12026.05 | 98 | 0.98 | |
| CF-GNN2026.05 | 94 | 3.18 | |
| SQR-GNN2026.05 | 78 | 0.53 | |
| RQRadj.-GNN2026.05 | 72 | 0.48 | |
| MC Dropout2026.05 | 33 | 0.16 |