Research map / Probabilistic and scientific modeling / Bayesian optimization
Are Random Decompositions all we need in High Dimensional Bayesian Optimisation?
Juliusz Krzysztof Ziomek, Haitham Bou Ammar
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 |
|---|---|---|---|
| Use random tree decomposition sampling instead of data-driven decomposition learning | High-dimensional BO on the reported benchmark suite · Synthetic functions, NAS, MIP, and LassoBench | best regret | Improved |
| Use random tree decomposition sampling instead of Tree's learned decomposition | RDUCB versus Tree · Low-dimensional benchmark settings | final best regret | Mixed |
| Use random tree decomposition sampling instead of Tree's learned decomposition | RDUCB versus Tree · Increasing-dimensional Stybtang and LassoBench DNA | final best regret | Improved |
| Add random decomposition sampling to HEBO | HEBO · Bayesmark MLP-Adam | normalized score: 92.68 ± 0.22 -> 93.67 ± 0.30 | Improved |
| Add random decomposition sampling to HEBO | HEBO · Bayesmark MLP-SGD | normalized score: 90.66 ± 0.81 -> 91.65 ± 0.10 | Improved |
| Add random decomposition sampling to HEBO | HEBO · Bayesmark DT | normalized score: 79.42 ± 0.45 -> 80.79 ± 0.15 | Improved |
| Add random decomposition sampling to HEBO | HEBO · Bayesmark RF | normalized score: 84.97 ± 0.32 -> 87.64 ± 2.00 | Improved |
| Add random decomposition sampling to HEBO | HEBO · Bayesmark average | normalized score: 86.93 ± 0.45 -> 88.44 ± 0.64 | Improved |
| Use uniform random sampling over tree decompositions | Data-independent decomposition selection · Tree-decomposition class under the stated adversarial assumptions | expected cumulative function mismatch | Improved |
Closest papers by meaning
- Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization (ICML 2025 · 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 0.88)
- This Too Shall Pass: Removing Stale Observations in Dynamic Bayesian Optimization (NeurIPS 2024 · similarity 0.88)
- Understanding High-Dimensional Bayesian Optimization (ICML 2025 · similarity 0.88)
- BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability (NeurIPS 2026 · similarity 0.88)
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.
