Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Bayesian Active Learning with Fully Bayesian Gaussian Processes
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 |
|---|---|---|---|
| use B-QBC (committee from hyperparameter posterior samples) instead of ALM | FBGP active learning · 6 simulators overall | NLML relative AUC decrease: 0.0 -> 41.3 | Helped |
| use QB-MGP (B-ALM + B-QBC variance of mixture) instead of ALM | FBGP active learning · 6 simulators overall | RMSE relative AUC decrease: 0.0 -> 12.0 | Helped |
| use B-QBC instead of ALM, RMSE | FBGP active learning · 6 simulators overall | RMSE relative AUC decrease: 0.0 -> 9.2 | Helped |
| use BALD instead of ALM | FBGP active learning · 6 simulators overall | RMSE relative AUC decrease: 0.0 -> -48.6 | Hurt |
| use BALD instead of ALM, NLML | FBGP active learning · 6 simulators overall | NLML relative AUC decrease: 0.0 -> -8.0 | Mixed |
| use Bayesian B-ALM (average over posterior samples) instead of ALM | FBGP active learning · 6 simulators overall | NLML relative AUC decrease: 0.0 -> 16.1 | Helped |
| add B-ALM variance term to B-QBC to form QB-MGP | FBGP active learning · Gramacy2d | RMSE relative AUC decrease: 5.9 -> 8.8 | Helped |
| add B-ALM variance term to B-QBC to form QB-MGP | FBGP active learning · Motorcycle (heteroscedastic noise) | performance curve | Helped |
| QB-MGP replaces B-QBC for NLML | FBGP active learning · 6 simulators overall | NLML relative AUC decrease: 41.3 -> 29.3 | Hurt |
Closest papers by meaning
- Self-Correcting Bayesian Optimization through Bayesian Active Learning (NeurIPS 2023 · similarity 0.88)
- Efficient Active Learning for Gaussian Process Classification by Error Reduction (NeurIPS 2021 · similarity 0.87)
- Bayesian Batch Active Learning as Sparse Subset Approximation (NeurIPS 2019 · similarity 0.87)
- Objective Bound Conditional Gaussian Process for Bayesian Optimization (ICML 2021 · similarity 0.85)
- Active Learning for Gaussian Process Regression Under Self-Induced Boltzmann Weights (NeurIPS 2026 · similarity 0.85)
- 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.