Research map / Probabilistic and scientific modeling / Bayesian optimization
Bridging Spherical Black-Box Optimizers
Johannes Ackermann, Stefano Peluchetti
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 |
|---|---|---|---|
| use ES-OVI instead of plain ES or OVI | BBOB 2D · BBOB 2D tasks (Attractive Sector) | fitness | Improved |
| use ES-OVI instead of plain ES or OVI | BBOB 2D · BBOB 2D tasks (GriewankRosebrock) | fitness | Hurt |
| use intermediate alpha=0.75 in ES-OVI instead of ES (0.0) or OVI (1.0) | ES-OVI · Brax locomotion tasks | reward | Improved |
| use ES-OVI (alpha=0.5) instead of ES/OVI | ES-OVI · Brax locomotion tasks | reward | Mixed |
| increase OVI weight alpha in ES-OVI (toward OVI) | ES-OVI · Acrobot | sensitivity to disturbances | Hurt |
| use intermediate alpha instead of ES or OVI | ES-OVI · Acrobot | robustness to strong action/observation noise | Improved |
| use AdaPol (CH plus cCBO) instead of other CBO variants | AdaPol · Brax locomotion tasks | reward | Improved |
| use cCBO instead of OVI with good initialization | cCBO · Model merging (MGSM test) | test accuracy | Improved |
| use cCBO instead of OVI with bad initialization | cCBO · Model merging (MGSM test) | test accuracy | Hurt |
| use SchedPol/AdaPol hybrid instead of OVI or cCBO alone | SchedPol/AdaPol · Model merging (MGSM test) | test accuracy | Improved |
| use ES-OVI instead of OpenAI-ES (runtime) | ES-OVI · Runtime, D=5 | runtime (s): 4.0s -> 5.8s | Hurt |
| use AdaPol instead of OVI (runtime) | AdaPol · Runtime, D=5 | runtime (s): 4.1s -> 13.8s | Hurt |
Closest papers by meaning
- PyPop7: A Pure-Python Library for Population-Based Black-Box Optimization (JMLR 2024 · similarity 0.88)
- MetaBox-v2: A Unified Benchmark Platform for Meta-Black-Box Optimization (NeurIPS 2025 · similarity 0.86)
- Optimizing the Unknown: Black Box Bayesian Optimization with Energy-Based Model and Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- DesignX: Human-Competitive Algorithm Designer for Black-Box Optimization (NeurIPS 2025 · similarity 0.86)
- Guided Trajectory Generation with Diffusion Models for Offline Model-based Optimization (NeurIPS 2024 · similarity 0.84)
- ConfigX: Modular Configuration for Evolutionary Algorithms via Multitask Reinforcement Learning (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.
