Research map / Probabilistic and scientific modeling / Bayesian optimization
BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces
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 |
|---|---|---|---|
| BASC Bayesian optimization instead of default differential evolution | CO on stoichiometric hematite · CO on stoichiometric hematite | DFT calls to within 0.1 eV (median): 192 -> 100 | Helped |
| BASC instead of default DE, best case | CO on stoichiometric hematite · CO on stoichiometric hematite | DFT calls to within 0.1 eV (best): 181 -> 94 | Helped |
| aggressive DE settings instead of default DE (ablation) | Differential evolution · CO on stoichiometric hematite | DFT calls to within 0.1 eV: 181 -> 93 | Mixed |
| BASC instead of DE on oxygen-deficient hematite | CO on oxygen-deficient hematite · CO on oxygen-deficient hematite | DFT calls to within 0.1 eV: 198 -> 96 | Helped |
| add unconstrained L-BFGS relaxation after BASC (ablation) | BASC · CO on stoichiometric hematite | energy (eV) / extra DFT calls: -188.81 eV with 26 more DFT calls | Helped |
| add L-BFGS relaxation after BASC on defect surface (ablation) | BASC · CO on oxygen-deficient hematite | energy (eV): -180.74 vs -180.46 | Helped |
| allow deformation by adsorption (CMH) versus rigid parameterized objective | CMH · CO on stoichiometric hematite | C-O bond length (A): 1.128 -> 1.141 | Mixed |
| noise variance above about 1e-4 (ablation) | BASC GP · CO on hematite | search behavior | Hurt |
| aggressive DE settings on defect surface | Differential evolution · CO on oxygen-deficient hematite | stability | Hurt |
| exploration/exploitation scheduling of output variance | BASC · CO on hematite | convergence | Helped |
Closest papers by meaning
- AdsorbDiff: Adsorbate Placement via Conditional Denoising Diffusion (ICML 2024 · similarity 0.85)
- A Sober Look at LLMs for Material Discovery: Are They Actually Good for Bayesian Optimization Over Molecules? (ICML 2024 · similarity 0.85)
- A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences (NeurIPS 2024 · similarity 0.84)
- Approximation-Aware Bayesian Optimization (NeurIPS 2024 · similarity 0.84)
- Empower Structure-Based Molecule Optimization with Gradient Guided Bayesian Flow Networks (ICML 2025 · similarity 0.84)
- De Novo Molecular and Crystal Design with Latent Space Bayesian Optimization (AAAI 2025 · similarity 0.83)
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.