Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Gaussian Process Boosting
Fabio Sigrist
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 |
|---|---|---|---|
| use GPBoost instead of a linear mixed effects model | GPBoost · Simulated grouped data (hajjem) | RMSE: 1.342 -> 1.100 | Improved |
| use GPBoost instead of LightGBM-style boosting with the group as categorical variable | GPBoost · Simulated grouped data (hajjem) | RMSE: 1.156 -> 1.100 | Improved |
| use GPBoost instead of MERF | GPBoost · Simulated grouped data (hajjem) | RMSE: 1.104 -> 1.100 | Improved |
| use GPBoost instead of CatBoost for new groups | GPBoost · Simulated grouped data (hajjem) | RMSE (new groups): 1.464 -> 1.458 | Improved |
| estimate covariance parameters on out-of-sample data (GPBoostOOS) | GPBoost · Simulated grouped data (hajjem) | RMSE of error variance: 0.1703 -> 0.1375 | Improved |
| use GPBoost instead of MERF for training time | GPBoost · Simulated grouped data (hajjem) | training time (s): 416.7 -> 0.9356 | Improved |
| use GPBoost instead of independent boosting with coordinates as predictors | GPBoost · Simulated spatial data (hajjem) | RMSE: 1.474 -> 1.374 | Improved |
| use GPBoost instead of a linear Gaussian process | GPBoost · Simulated spatial data (hajjem) | RMSE: 1.466 -> 1.374 | Improved |
| use GPBoost instead of a two-step approach | GPBoost · Simulated spatial data (hajjem) | RMSE: 1.406 -> 1.374 | Improved |
| use GPBoost instead of independent boosting for extrapolation | GPBoost · Simulated spatial data (hajjem) | RMSE (extrapolation): 1.611 -> 1.519 | Improved |
| use GPBoost instead of independent boosting for sums | GPBoost · Simulated spatial data (hajjem) | CRPS sum: 8.508 -> 6.468 | Improved |
| use GPBoost instead of a linear model when the true function is linear | GPBoost · Simulated data (linear F) | RMSE | Hurt |
Closest papers by meaning
- Latent Gaussian Model Boosting (TPAMI 2022 · similarity 0.91)
- Gradient Boosting Performs Gaussian Process Inference (ICLR 2023 · similarity 0.87)
- Generic Inference in Latent Gaussian Process Models (JMLR 2019 · similarity 0.87)
- Minimizing Negative Transfer of Knowledge in Multivariate Gaussian Processes: A Scalable and Regularized Approach (TPAMI 2020 · similarity 0.86)
- Additive Gaussian Processes Revisited (ICML 2022 · similarity 0.86)
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.85)
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.
