Node-level Regression Uncertainty Quantification on Tree
100PICPER-GNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ER-GNN2026.05 | 100 | 11.21 | |
| BayesianNN2026.05 | 100 | 3 | |
| QpiGNNWidth penalty (lambda)=0.12026.05 | 100 | 0.59 | |
| QpiGNNWidth penalty (lambda)=0.52026.05 | 96 | 0.39 | |
| QpiGNNWidth penalty (lambda)=opt.2026.05 | 96 | 0.39 | |
| CF-GNN2026.05 | 93 | 0.97 | |
| RQRadj.-GNN2026.05 | 85 | 0.68 | |
| SQR-GNN2026.05 | 80 | 0.26 | |
| MC Dropout2026.05 | 64 | 0.2 |