Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Uncertainty in Gradient Boosting via Ensembles
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Estimate knowledge uncertainty with an ensemble instead of total uncertainty | GBDT uncertainty estimates · Classification and regression datasets | OOD AUC-ROC | Helped |
| Use an ensemble rather than a single GBDT for total uncertainty | GBDT uncertainty estimates · Classification and regression datasets | error-detection PRR | No clear effect |
| Use virtual SGLB samples from one boosting trajectory rather than independent models (ablation) | SGLB uncertainty estimates · Classification and regression datasets | OOD AUC-ROC: 89 vs 85 | Hurt |
| Use virtual SGLB rather than independent SGLB models (ablation) | SGLB uncertainty estimates · Classification datasets with categorical features | OOD AUC-ROC | Helped |
| Use total uncertainty rather than knowledge uncertainty | GBDT error ranking · Classification and regression datasets | error-detection PRR | Helped |
| Use SGLB ensembles rather than SGB ensembles | GBDT uncertainty estimation · Classification and regression datasets | OOD AUC-ROC | No clear effect |
| Add ensemble averaging to a single SGB model | CatBoost GBDT · Adult | classification NLL: 0.276 -> 0.271 | Helped |
| Add ensemble averaging to a single SGB model | CatBoost GBDT · BostonH | regression RMSE: 3.06 -> 3.04 | No clear effect |
Closest papers by meaning
- Neural Ensemble Search for Uncertainty Estimation and Dataset Shift (NeurIPS 2021 · similarity 0.88)
- Wasserstein Gradient Boosting: A Framework for Distribution-Valued Supervised Learning (NeurIPS 2024 · similarity 0.87)
- Theoretical Limitations of Ensembles in the Age of Overparameterization (ICML 2025 · similarity 0.87)
- Structure-informed Risk Minimization for Robust Ensemble Learning (ICML 2025 · similarity 0.87)
- Statistical Inference for Gradient Boosting Regression (NeurIPS 2025 · similarity 0.87)
- Free Lunch in the Forest: Functionally-Identical Pruning of Boosted Tree Ensembles (AAAI 2025 · similarity 0.86)
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.