Regression on Synthetic two-feature data Sin n=1000
0.9954Mean R^2LightGBM, rel-s features
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LightGBM, rel-s featuresSample size (n)=10002025.12 | 0.9954 | 0.0008 | |
| NW + relationsNorm/Embedding Type=L2 norm, Sample size (n)=10002025.12 | 0.991 | 0.0011 | |
| NW + relationsNorm/Embedding Type=MLP embeddings, Sample size (n)=10002025.12 | 0.9822 | 0.0025 | |
| NW + relationsNorm/Embedding Type=learnable norm, Sample size (n)=10002025.12 | 0.9774 | 0.0019 | |
| TabRelSample size (n)=10002025.12 | 0.9392 | 0.1388 | |
| NW, no rel-s infoNorm/Embedding Type=learnable norm, Sample size (n)=10002025.12 | 0.7747 | 0.016 | |
| NW, no rel-s infoNorm/Embedding Type=L2 norm, Sample size (n)=10002025.12 | 0.7729 | 0.0161 | |
| LightGBM, no rel-s infoSample size (n)=10002025.12 | 0.7361 | 0.0219 | |
| NW, rel-s featuresSample size (n)=10002025.12 | -0.3593 | 0.074 |