Research map / Probabilistic and scientific modeling / Bayesian optimization
Bayesian Optimization of Combinatorial Structures
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 BOCS-SDP rather than EI | BQP with Lc=100 and λ=0 · BQP, 10 variables, 100 iterations | simple regret (×10): 3.38 ± 0.70 -> 0.11 ± 0.06 | Helped |
| Use BOCS-SDP rather than BOCS-SA | Ising sparsification · Ising, λ=10^-4, 150 iterations | best objective value: 0.19 ± 0.04 -> 0.10 ± 0.03 | Helped |
| Use BOCS-SDP rather than EI | Contamination control · Contamination control, λ=10^-2, 250 iterations | best objective value: 21.54 ± 0.01 -> 21.48 ± 0.01 | Helped |
| Replace the SDP acquisition optimizer with simulated annealing (ablation) | BOCS · BQP, 10 variables, Lc=100, λ=0 | simple regret (×10): 0.11 ± 0.06 -> 0.15 ± 0.07 | Hurt |
| Replace posterior sampling of regression weights with an MLE point estimate (ablation) | BOCS-SA · BQP | simple regret | Hurt |
| Increase the model order from second to third order (ablation) | BOCS-SA on the aero-structural benchmark · Aero-structural, λ=10^-2 | objective value | Mixed |
| Increase the model order from second to third order (ablation) | BOCS-SA on Ising sparsification · Ising, λ=0 and λ=10^-2 | objective value | Helped |
| Use sparse Bayesian regression instead of standard Bayesian linear regression (ablation) | BOCS regression model · Ising model test set | absolute approximation error | Helped |
Closest papers by meaning
- Bayesian Optimization of Functions over Node Subsets in Graphs (NeurIPS 2024 · similarity 0.91)
- Mercer Features for Efficient Combinatorial Bayesian Optimization (AAAI 2021 · similarity 0.90)
- Omnipresent Yet Overlooked: Heat Kernels in Combinatorial Bayesian Optimization (NeurIPS 2025 · similarity 0.89)
- Relaxing the Additivity Constraints in Decentralized No-Regret High-Dimensional Bayesian Optimization (ICLR 2024 · similarity 0.87)
- A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences (NeurIPS 2024 · similarity 0.87)
- Adaptive Partitioning Schemes for Optimistic Optimization (ICML 2025 · 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.