Research map / Probabilistic and scientific modeling / Bayesian optimization
Lookahead Bayesian Optimization with Inequality Constraints
Remi Lam, Karen Willcox
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 |
|---|---|---|---|
| rollout h=1 instead of greedy EIC | constrained BO · P1 | log10 median utility gap: -4.45 -> -4.59 | Improved |
| rollout h=1 instead of EIC | constrained BO · P2 | log10 median utility gap: -2.62 -> -2.99 | Improved |
| rollout h=1 instead of EIC | constrained BO · P3 | log10 median utility gap: 1.60 -> 1.48 | Improved |
| rollout h=1 instead of EIC | constrained BO · P4 | log10 median utility gap: 0.57 -> -0.10 | Improved |
| rollout instead of PESC | constrained BO · P1 | log10 median utility gap: -2.68 -> -4.59 | Improved |
| longer rolling horizon h=3 vs h=1 (ablation) | rollout · P3 | log10 median utility gap: 1.48 -> 1.35 | Improved |
| longer rolling horizon h=2 vs h=1 (ablation) | rollout · P3 | log10 median utility gap: 1.48 -> 1.31 | Improved |
| longer rolling horizon h=3 vs h=1 (ablation) | rollout · P1 | log10 median utility gap: -4.59 -> -4.42 | Hurt |
| longer rolling horizon h=3 vs h=1 (ablation) | rollout · P4 | log10 median utility gap: -0.10 -> 0.19 | Hurt |
| longer rolling horizon h=3 vs h=1 (ablation) | rollout · P2 | log10 median utility gap: -2.99 -> -2.994 | No clear effect |
| pure exploitation (PM) vs lookahead | PM · P1 | log10 median utility gap: 0.30 -> -4.59 | Hurt |
| local optimizers (SLSQP/MMA/COBYLA) vs rollout | local algorithms · P3 | log10 median utility gap: 3.06 -> 1.48 | Hurt |
Closest papers by meaning
- Hypervolume Knowledge Gradient: A Lookahead Approach for Multi-Objective Bayesian Optimization with Partial Information (ICML 2023 · similarity 0.91)
- Optimistic Bayesian Optimization with Unknown Constraints (ICLR 2024 · similarity 0.90)
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.89)
- Preference-Aware Constrained Multi-Objective Bayesian Optimization (Student Abstract) (AAAI 2024 · similarity 0.88)
- MOBO-OSD: Batch Multi-Objective Bayesian Optimization via Orthogonal Search Directions (NeurIPS 2025 · similarity 0.87)
- Constrained Bayesian Optimization under Partial Observations: Balanced Improvements and Provable Convergence (AAAI 2024 · similarity 0.87)
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.
