Research map / Probabilistic and scientific modeling / Bayesian optimization
Framework and Benchmarks for Combinatorial and Mixed-variable Bayesian Optimization
Kamil Dreczkowski, Antoine Grosnit, 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 a dynamic trust region for local modeling and acquisition optimization (ablation) | Combinatorial BO · Six combinatorial tasks | average rank | Improved |
| Use a dynamic trust region for local modeling and acquisition optimization (ablation) | Mixed-variable BO · Four mixed-variable tasks | average rank | Improved |
| Use GP (HED) rather than GP (O) or GP (TO) (ablation) | Mixed-variable BO · Four mixed-variable tasks | average rank | Improved |
| Use LSH or GP (Diff.) rather than GP (HED), GP (O), GP (TO), or GP (SSK) (ablation) | Combinatorial BO · Six combinatorial tasks | average rank | Hurt |
| Use exhaustive Local Search rather than GA (ablation) | Combinatorial BO · Low evaluation budgets | average rank | Improved |
| Use GA rather than exhaustive Local Search (ablation) | Combinatorial BO · Increasing evaluation budget | average rank | Improved |
| Use MAB-GD rather than the other tested acquisition optimizers (ablation) | Mixed-variable BO · Four mixed-variable tasks | average rank | Hurt |
| Use HC-GD or SA rather than GA (ablation) | Mixed-variable BO · Low evaluation budgets | average rank | Improved |
| Use HC-GD or SA rather than GA (ablation) | Mixed-variable BO · High evaluation budgets | average rank | No clear effect |
| Combine GP (SSK) or GP (TO) with GA and a trust region | Combinatorial BO · Six combinatorial tasks | average rank | Improved |
| Combine GP (Matérn-5/2 + HED) with GA and a trust region | Mixed-variable BO · Four mixed-variable tasks | average rank | Improved |
| Use a non-BO optimizer instead of BO | Combinatorial optimization · Antibody design and RNA inverse folding | objective value | Mixed |
Closest papers by meaning
- A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences (NeurIPS 2024 · similarity 0.88)
- A General Framework for User-Guided Bayesian Optimization (ICLR 2024 · similarity 0.88)
- Omnipresent Yet Overlooked: Heat Kernels in Combinatorial Bayesian Optimization (NeurIPS 2025 · similarity 0.87)
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.87)
- Understanding High-Dimensional Bayesian Optimization (ICML 2025 · similarity 0.87)
- BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability (NeurIPS 2026 · 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.
