Research map / Probabilistic and scientific modeling / Bayesian optimization
Tree ensemble kernels for Bayesian optimization with known constraints over mixed-feature spaces
Alexander Thebelt, Calvin Tsay, Robert Matthew Lee, Nathan Sudermann-Merx, David Walz, Behrang Shafei, Ruth Misener
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 |
|---|---|---|---|
| increase allowed tree-agreement ratio R from 0.35 to 1.0 (ablation) | LEAF-GP uncertainty constraint · Rastrigin 10D | model error | Improved |
| replace Gurobi acquisition optimization with random sampling (LEAF-GP-RND) (ablation) | LEAF-GP · Hartmann 6D | best objective | No clear effect |
| replace Gurobi acquisition optimization with random sampling (LEAF-GP-RND) (ablation) | LEAF-GP · Styblinski-Tang 10D | best objective | Hurt |
| use LEAF-GP instead of Matern GP BO | UCB/EI-MATERN · Hartmann 6D | best objective | Hurt |
| use LEAF-GP instead of other baselines | BO baselines · Styblinski-Tang 10D | best objective | Improved |
| use LEAF-GP with explicit constraints instead of baselines and FEAS-RANDOM | BO baselines · G3, G4, Alkylation | feasible best objective | Improved |
| use LEAF-GP instead of UCB-MATERN | BO baselines · G1 | feasible best objective | Mixed |
| use LEAF-GP instead of other methods | BO baselines · Pressure Vessel 4D | feasible best objective | Improved |
| tree-based models instead of Matern GPs | BO surrogate · CIFAR-NAS 29D | test accuracy | Improved |
| LEAF-GP instead of SMAC | BO · CIFAR-NAS 29D | test accuracy | Hurt |
| use LEAF-GP instead of other algorithms | BO · VAE-NAS | feasible best objective | Improved |
Closest papers by meaning
- BARK: A Fully Bayesian Tree Kernel for Black-box Optimization (ICML 2025 · similarity 0.89)
- Neural Tangent Kernels for Axis-Aligned Tree Ensembles (ICML 2024 · similarity 0.87)
- Minimax Rates and Spectral Distillation for Tree Ensembles (NeurIPS 2026 · similarity 0.87)
- Compressing tree ensembles through Level-wise Optimization and Pruning (ICML 2025 · similarity 0.86)
- Adaptive Kernel Design for Bayesian Optimization Is a Piece of CAKE with LLMs (NeurIPS 2025 · similarity 0.85)
- Automated Kernel Discovery Towards Understanding High-dimensional Bayesian Optimization (NeurIPS 2026 · similarity 0.84)
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.
