Classification on HE 50% corrupted features noise
39.3AccuracyAGL + FT
Evaluation Results
| Method | Links | |
|---|---|---|
| AGL + FTFeature Selection=AGL, Downstream Model=FT-Transformer2026.03 | 39.3 | |
| Deep Lasso + FTFeature Selection=Deep Lasso, Downstream Model=FT-Transformer2026.03 | 39.3 | |
| 1L Lasso + FTFeature Selection=1L Lasso, Downstream Model=FT-Transformer2026.03 | 39.2 | |
| LassoNet + FTFeature Selection=LassoNet, Downstream Model=FT-Transformer2026.03 | 39.2 | |
| AM + FTFeature Selection=AM, Downstream Model=FT-Transformer2026.03 | 39.2 | |
| XGBoost + FTFeature Selection=XGBoost, Downstream Model=FT-Transformer2026.03 | 39.2 | |
| LassoFlexNetFeature Selection=End-to-end2026.03 | 39 | |
| Lasso + FTFeature Selection=Lasso, Downstream Model=FT-Transformer2026.03 | 38.9 | |
| Deep Lasso + MLPFeature Selection=Deep Lasso, Downstream Model=MLP2026.03 | 38.8 | |
| Univariate + FTFeature Selection=Univariate, Downstream Model=FT-Transformer2026.03 | 38.7 | |
| RF + FTFeature Selection=RF, Downstream Model=FT-Transformer2026.03 | 38.7 | |
| AGL + MLPFeature Selection=AGL, Downstream Model=MLP2026.03 | 38.6 | |
| XGBoost + MLPFeature Selection=XGBoost, Downstream Model=MLP2026.03 | 38.5 | |
| RF + MLPFeature Selection=RF, Downstream Model=MLP2026.03 | 38.3 | |
| Lasso + MLPFeature Selection=Lasso, Downstream Model=MLP2026.03 | 38.2 | |
| 1L Lasso + MLPFeature Selection=1L Lasso, Downstream Model=MLP2026.03 | 38.2 | |
| LassoNet + MLPFeature Selection=LassoNet, Downstream Model=MLP2026.03 | 38.2 | |
| AM + MLPFeature Selection=AM, Downstream Model=MLP2026.03 | 38.1 | |
| No FS + MLPFeature Selection=None, Downstream Model=MLP2026.03 | 37 | |
| Univariate + MLPFeature Selection=Univariate, Downstream Model=MLP2026.03 | 34.6 | |
| No FS + FTFeature Selection=None, Downstream Model=FT-Transformer2026.03 | 34.5 |