Research map / Probabilistic and scientific modeling / Bayesian optimization
BOCK : Bayesian Optimization with Cylindrical Kernels
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 BOCK instead of the strongest listed baseline, Matern | Bayesian optimization · Repeated Hartmann6, 100 dimensions | minimum objective: -2.91$\pm$0.26 -> -3.16$\pm$0.10 | Helped |
| use BOCK instead of Matern | Bayesian optimization · Rosenbrock, 20 and 100 dimensions | minimum objective: 230.25$\pm$ 187.41 -> 47.87$\pm$ 33.94; 231.42$\pm$ 28.94 -> 128.69$\pm$ 52.84 | Helped |
| use BOCK instead of Matern | Bayesian optimization · Repeated Branin, 20 and 100 dimensions | minimum objective: 0.41$\pm$0.00 -> 0.50$\pm$0.12; 0.54$\pm$0.06 -> 1.03$\pm$0.17 | Mixed |
| use BOCK instead of Matern | Bayesian optimization · Levy, 20 and 100 dimensions | minimum objective: 0.38$\pm$0.13 -> 0.54$\pm$0.13; 2.17$\pm$0.18 -> 6.78$\pm$2.16 | Hurt |
| optimize stochastic-depth death rates with BOCK rather than using linear death rates | SDResNet-110 · CIFAR100 test | accuracy: 74.90$\pm$0.15 -> 75.06$\pm$0.19 | Helped |
| optimize stochastic-depth death rates with BOCK rather than using linear death rates | SDResNet-110 · CIFAR100 test | expected depth: ~~82.50 -> 74.51$\pm$1.22 | Helped |
| add radius input warping (ablation) | BOCK · Rosenbrock, 20 dimensions | minimum objective: 1314.03$\pm$ 1619.73 -> 47.87$\pm$ 33.94 | Helped |
| add radius input warping (ablation) | BOCK · Rosenbrock, 50 dimensions | minimum objective: 51.14$\pm$58.18 -> 29.65$\pm$11.56 | Helped |
| add radius input warping (ablation) | BOCK · Rosenbrock, 100 dimensions | minimum objective: 157.89$\pm$ 161.92 -> 128.69$\pm$~~52.84 | Helped |
| add fake boundary data to reduce predictive variance (ablation) | BOCK · Rosenbrock, 20 dimensions | minimum objective: 47.87$\pm$~~~~ 33.94 -> ~~~~48.87$\pm$~~~~ 18.33 | No clear effect |
| add fake boundary data to reduce predictive variance (ablation) | BOCK · Rosenbrock, 50 dimensions | minimum objective: 29.65$\pm$11.56 -> 33.90$\pm$21.69 | Hurt |
| add fake boundary data to reduce predictive variance (ablation) | BOCK · Rosenbrock, 100 dimensions | minimum objective: 128.69$\pm$~~52.84 -> ~~87.00$\pm$~~36.88 | Helped |
Closest papers by meaning
- Standard Gaussian Process is All You Need for High-Dimensional Bayesian Optimization (ICLR 2025 · similarity 0.89)
- BOIDS: High-Dimensional Bayesian Optimization via Incumbent-Guided Direction Lines and Subspace Embeddings (AAAI 2025 · similarity 0.89)
- Joint Composite Latent Space Bayesian Optimization (ICML 2024 · similarity 0.89)
- Minimizing UCB: a Better Local Search Strategy in Local Bayesian Optimization (NeurIPS 2024 · similarity 0.89)
- Vanilla Bayesian Optimization Performs Great in High Dimensions (ICML 2024 · similarity 0.89)
- Automated Kernel Discovery Towards Understanding High-dimensional Bayesian Optimization (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.