Regression on Boston (NLL, RMSE)
2.65RMSECDropout
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CDropout2024.05 | 2.65 | 2.72 | |
| WEvidential2024.05 | 2.78 | 2.47 | |
| GPExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=322026.07 | 2.83 | — | |
| GPExperimental Setting=2, Learning Rate Setting=low2026.07 | 2.83 | 2.41 | |
| MFVIExperimental Setting=2, Learning Rate Setting=low2026.07 | 2.89 | 3.27 | |
| VBLLExperimental Setting=2, Learning Rate Setting=low2026.07 | 2.92 | 3.25 | |
| NGBoost2024.05 | 2.94 | 2.43 | |
| MCDropout2024.05 | 2.97 | 2.46 | |
| DEExperimental Setting=2, Learning Rate Setting=low2026.07 | 2.99 | 2.52 | |
| BDKNExperimental Setting=2, Learning Rate Setting=low2026.07 | 3.01 | 2.81 | |
| BDKNExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons), Feature maps (H)=52026.07 | 3.03 | — | |
| DEvidential2024.05 | 3.06 | 2.35 | |
| BLLExperimental Setting=2, Learning Rate Setting=low2026.07 | 3.2 | 3.21 | |
| DEnsemble2024.05 | 3.28 | 2.41 | |
| DKLExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 3.45 | — | |
| DEExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 3.6 | — | |
| VBLLExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 3.68 | — | |
| BLLExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 4.04 | — | |
| DKLExperimental Setting=2, Learning Rate Setting=low2026.07 | 4.11 | 13.05 | |
| LDBLLExperimental Setting=2, Learning Rate Setting=low2026.07 | 4.41 | 3.3 | |
| MFVIExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 5.45 | — | |
| LDBLLExperimental setting=1 (high learning rate), Learning rate=0.1, Training epochs=40, Batch size=32, Architecture=2 hidden layers (50 neurons)2026.07 | 7.8 | — |