Node-level Regression Uncertainty Quantification on Chameleon
100PICPBayesianNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BayesianNN2026.05 | 100 | 2.95 | |
| QpiGNNWidth penalty (lambda)=0.12026.05 | 98 | 0.4 | |
| ER-GNN2026.05 | 97 | 1.08 | |
| QpiGNNWidth penalty (lambda)=opt.2026.05 | 96 | 0.23 | |
| RQRadj.-GNN2026.05 | 86 | 0.15 | |
| QpiGNNWidth penalty (lambda)=0.52026.05 | 51 | 0.03 | |
| MC Dropout2026.05 | 47 | 0.02 | |
| SQR-GNN2026.05 | 37 | 0.01 |