Research map / Probabilistic and scientific modeling / Bayesian optimization
Incorporating Expert Priors into Bayesian Optimization via Dynamic Mean Decay
Chongqi Qu, Meiqin Liu, Jian Lan, Shanling Dong, Zhunga Liu
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 |
|---|---|---|---|
| add expert-prior mean with dynamic decay (good prior) | BO with PI, EI, LogEI, TS, UCB, KG, MES · Hartmann4, Levy5, Hartmann6, Rosenbrock6, Stybtang7, PD1 (LM1B, WMT, CIFAR) | best-so-far value | Improved |
| use DynMeanBO instead of pi-BO/ColaBO under a good prior | DynMeanBO-EI/LogEI/MES · synthetic and PD1 tasks (good prior) | best-so-far value | No clear effect |
| use DynMeanBO instead of pi-BO/ColaBO under a good prior | DynMeanBO · synthetic and PD1 tasks (good prior) | per-iteration time | Improved |
| use DynMeanBO instead of pi-BO/ColaBO under a bad prior | DynMeanBO · synthetic and PD1 tasks (bad prior) | best-so-far value | Improved |
| add decaying expert-prior mean on PD1 LM1B under bad prior | DynMeanBO, piBO, ColaBO · PD1 LM1B (bad prior) | best-so-far value | Improved |
| add expert-prior mean (computational overhead) | DynMeanBO vs standard BO · synthetic and HPO tasks | per-iteration time | No clear effect |
| vary scale A of prior mean (ablation) | DynMeanBO-EI · eight tasks (good prior) | best-so-far value | No clear effect |
| vary shift B of prior mean (ablation) | DynMeanBO-EI · eight tasks (good prior) | best-so-far value | No clear effect |
| add DynMeanBO under 20-dimensional tasks | BO with seven acquisition functions · Levy 20D, Rosenbrock 20D | best-so-far value | Improved |
Closest papers by meaning
- A General Framework for User-Guided Bayesian Optimization (ICLR 2024 · similarity 0.91)
- Large Language Models to Enhance Bayesian Optimization (ICLR 2024 · similarity 0.89)
- Understanding High-Dimensional Bayesian Optimization (ICML 2025 · similarity 0.89)
- BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability (NeurIPS 2026 · similarity 0.89)
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.89)
- Vanilla Bayesian Optimization Performs Great in High Dimensions (ICML 2024 · similarity 0.89)
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.
