Research map / Probabilistic and scientific modeling / Bayesian optimization
Modulating Surrogates for Bayesian Optimization
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 latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Corrupted Holder Table, 50 evals | mean gap: 0.826 -> 0.896 | Helped |
| Use latent-input modulation instead of a GP | Bayesian optimization with GP surrogates · Shubert, 50 evals | mean gap: 0.378 -> 0.593 | Helped |
| Use latent-input modulation instead of a GP | Bayesian optimization with GP surrogates · Holder Table, 50 evals | mean gap: 0.939 -> 0.993 | Helped |
| Use latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Ackley [-10, 30]^d, 100 evals, 6D | mean gap: 0.789 -> 0.712 | Hurt |
| Use latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Weierstrass, 100 evals | mean gap: 0.704 -> 0.625 | Hurt |
| Use latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Deflected Corrugated Spring, 100 evals | mean gap: 0.840 -> 0.697 | Hurt |
| Use latent-input modulation instead of a GP | Bayesian optimization with GP surrogates · Shubert [-10, 10]^d, 100 evals | mean gap: 0.510 -> 0.877 | Helped |
| Use latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Corrupted Holder Table, 100 evals | mean gap: 0.889 -> 0.918 | Helped |
| Use latent-input modulation instead of a homoscedastic GP | Bayesian optimization with GP surrogates · Corrupted Exponential, 100 evals | mean gap: 0.847 -> 0.806 | Hurt |
| Use latent-input modulation instead of a heteroscedastic GP | Bayesian optimization with GP surrogates · Active learning: Robot Pushing, 100 evals | mean gap: 0.907 -> 0.932 | Helped |
| Use latent-input modulation rather than the standard GP | Bayesian optimization with GP surrogates · Across the benchmark suite | mean gap | Mixed |
Closest papers by meaning
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.91)
- Bayesian Optimisation with Unknown Hyperparameters: Regret Bounds Logarithmically Closer to Optimal (NeurIPS 2024 · similarity 0.89)
- Understanding High-Dimensional Bayesian Optimization (ICML 2025 · similarity 0.89)
- Informed Initialization for Bayesian Optimization and Active Learning (NeurIPS 2025 · similarity 0.89)
- Vanilla Bayesian Optimization Performs Great in High Dimensions (ICML 2024 · similarity 0.88)
- Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization (ICML 2025 · 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.