Research map / Graph learning / GNN architectures
Does your graph need a confidence boost? Convergent boosted smoothing on graphs with tabular node features
Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| interleave graph propagation with boosting instead of applying propagation only after CatBoost training (ablation) | CatBoost/EBBS · House, County, VK, Avazu | RMSE: 0.54 -> 0.45; 1.25 -> 1.11; 6.96 -> 6.90; 0.1083 -> 0.1062 | Improved |
| interleave graph propagation with boosting instead of applying propagation only after CatBoost training (ablation) | CatBoost/EBBS · SLAP, DBLP | accuracy: 96.2 -> 96.3; 90.7 -> 91.3 | Mixed |
| interleave graph propagation with boosting instead of applying propagation only after CatBoost training (ablation) | CatBoost/EBBS · Coauthor-CS, Coauthor-Phy | accuracy: 94.6 -> 94.9; 96.4 -> 96.9 | Improved |
| reduce BGNN to one SGD step per iteration (ablation) | BGNN · House, County, VK, Avazu | RMSE: 0.50 -> 0.56; 1.26 -> 1.34; 6.95 -> 7.28; 0.1090 -> 0.110 | Hurt |
| remove end-to-end training (ablation) | BGNN · House, County, VK, Avazu | RMSE: 0.50 -> 0.51; 1.26 -> 1.33; 6.95 -> 7.07; 0.1090 -> 0.1095 | Mixed |
| use graph propagation with boosted tabular features | EBBS · OGB-ArXiv | accuracy: 67.0 -> 70.10 | Improved |
| use EBBS instead of the strongest reported non-EBBS method | EBBS · House, County, VK, Avazu | RMSE: 0.50 -> 0.45; 1.25 -> 1.11; 6.95 -> 6.90; 0.1083 -> 0.1062 | Improved |
| use EBBS instead of the strongest reported non-EBBS method | EBBS · SLAP, DBLP, Coauthor-CS, Coauthor-Phy | accuracy: 96.3 -> 96.3; 91.3 -> 91.3; 94.6 -> 94.9; 96.9 -> 96.9 | Mixed |
Closest papers by meaning
- Graph Sparsification via Mixture of Graphs (ICLR 2025 · similarity 0.85)
- No Metric to Rule Them All: Toward Principled Evaluations of Graph-Learning Datasets (ICML 2025 · similarity 0.85)
- Gradient Regularized Newton Boosting Trees with Global Convergence (NeurIPS 2026 · similarity 0.85)
- Statistical Inference for Gradient Boosting Regression (NeurIPS 2025 · similarity 0.84)
- Free Lunch in the Forest: Functionally-Identical Pruning of Boosted Tree Ensembles (AAAI 2025 · similarity 0.84)
- NRGBoost: Energy-Based Generative Boosted Trees (ICLR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
