Research map / Probabilistic and scientific modeling / Bayesian optimization
Truncated Variance Reduction: A Unified Approach to Bayesian Optimization and Level-Set Estimation
Ilija Bogunovic, Jonathan Scarlett, Andreas Krause, Volkan Cevher
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| incorporate pointwise sampling costs into the acquisition rule | TruVaR vs. GCHK · Lake data, varying cost | F1-score | Improved |
| allow sampling to mix between noise levels | TruVaR vs. fixed-noise GCHK · Synthetic LSE data, multiple noise levels | F1-score | Improved |
| use TruVaR rather than GCHK or STR | Lake data, unit cost · Lake data, unit-cost LSE | F1-score | No clear effect |
| use GCHK or STR rather than maximum variance sampling | LSE algorithms · Lake data, unit-cost LSE | F1-score | Improved |
| use TruVaR rather than ES or MRS in later rounds | Bayesian optimization algorithms · Synthetic BO data | regret | Improved |
| use TruVaR rather than EI or GP-UCB | Bayesian optimization algorithms · SVM on grid | validation error | Improved |
| restrict acquisition-function selection to points in the current potential-maximizer set (ablation) | TruVaR · Experiments discussed in the supplementary material | performance | No clear effect |
| set the truncation threshold to zero for pure variance reduction (ablation) | TruVaR · Experiments discussed in the supplementary material | performance | No clear effect |
| replace the shrinking potential-maximizer set with the full domain during updates (ablation) | TruVaR for online kernel-hyperparameter learning · SVM on grid | validation error | No clear effect |
Closest papers by meaning
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.87)
- This Too Shall Pass: Removing Stale Observations in Dynamic Bayesian Optimization (NeurIPS 2024 · similarity 0.87)
- Vanilla Bayesian Optimization Performs Great in High Dimensions (ICML 2024 · similarity 0.87)
- Understanding High-Dimensional Bayesian Optimization (ICML 2025 · similarity 0.86)
- Optimizing Posterior Samples for Bayesian Optimization via Rootfinding (ICLR 2025 · similarity 0.86)
- Sample-efficient Bayesian Optimisation Using Known Invariances (NeurIPS 2024 · 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.
