Research map / Probabilistic and scientific modeling / Bayesian optimization
Instance Generation for Meta-Black-Box Optimization Through Latent Space Reverse Engineering
Chen Wang, Yue-Jiao Gong, Zhiguang Cao, Zeyuan Ma
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 |
|---|---|---|---|
| train MetaBBO on Diverse-BBO instead of CoCo-BBOB | MetaBBO average · D_synthetic | normalized performance: 7.939e-01 -> 7.980e-01 | Improved |
| train MetaBBO on Diverse-BBO instead of MA-BBOB | MetaBBO average · D_synthetic | normalized performance: 7.843e-01 -> 7.980e-01 | Improved |
| train MetaBBO on Diverse-BBO instead of GP-BBOB | MetaBBO average · D_synthetic | normalized performance: 7.816e-01 -> 7.980e-01 | Improved |
| train SYMBOL on Diverse-BBO instead of GP-BBOB | SYMBOL · D_synthetic | normalized performance: 6.907e-01 -> 7.346e-01 | Improved |
| train GLEET on Diverse-BBO instead of MA-BBOB | GLEET · D_synthetic | normalized performance: 8.112e-01 -> 8.106e-01 | No clear effect |
| train MetaBBO on Diverse-BBO instead of baseline benchmarks | MetaBBO (4 approaches) · realistic sets (HPO, UAV, protein) | average performance gain (1/rank) | Improved |
| use MA-BBOB or GP-BBOB for MetaBBO training instead of CoCo-BBOB | MetaBBO average · D_synthetic | normalized performance: 7.939e-01 -> 7.843e-01 | Hurt |
| add autoencoder latent space analysis instead of PCA (ablation) | LSRE · GP search objective curve | search objective | Improved |
| add vectorized symbol set (X[i:j]) instead of simple symbol set (ablation) | LSRE · GP search objective curve | search objective | Improved |
| add cross-dimensional local search (ablation) | LSRE · GP search objective curve | search objective | Improved |
| full LSRE instead of GP-BBOB (all novel designs) (ablation) | GP backbone · GP search objective curve | search objective | Improved |
Closest papers by meaning
- MetaBox-v2: A Unified Benchmark Platform for Meta-Black-Box Optimization (NeurIPS 2025 · similarity 0.89)
- Meta-Black-Box-Optimization through Offline Q-function Learning (ICML 2025 · similarity 0.87)
- SYMBOL: Generating Flexible Black-Box Optimizers through Symbolic Equation Learning (ICLR 2024 · similarity 0.87)
- ConfigX: Modular Configuration for Evolutionary Algorithms via Multitask Reinforcement Learning (AAAI 2025 · similarity 0.86)
- Neural Exploratory Landscape Analysis for Meta-Black-Box-Optimization (ICLR 2025 · similarity 0.86)
- B2Opt: Learning to Optimize Black-box Optimization with Little Budget (AAAI 2025 · similarity 0.84)
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.
