Regression on Synthetic two-feature data Sin, n=300
0.9835R^2 (Mean)LightGBM, rel-s features
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LightGBM, rel-s featuresSample size (n)=3002025.12 | 0.9835 | 0.0025 | |
| NW + relationsNorm/Embedding Type=L2 norm, Sample size (n)=3002025.12 | 0.9777 | 0.0045 | |
| NW + relationsNorm/Embedding Type=MLP embeddings, Sample size (n)=3002025.12 | 0.9692 | 0.0093 | |
| NW + relationsNorm/Embedding Type=learnable norm, Sample size (n)=3002025.12 | 0.9532 | 0.0116 | |
| TabRelSample size (n)=3002025.12 | 0.9327 | 0.0571 | |
| NW, no rel-s infoNorm/Embedding Type=learnable norm, Sample size (n)=3002025.12 | 0.7474 | 0.0441 | |
| NW, no rel-s infoNorm/Embedding Type=L2 norm, Sample size (n)=3002025.12 | 0.7349 | 0.0454 | |
| LightGBM, no rel-s infoSample size (n)=3002025.12 | 0.7233 | 0.0537 | |
| NW, rel-s featuresSample size (n)=3002025.12 | 0.3754 | 0.1459 |