Research map / Learning theory and algorithms / Combinatorial and search optimization
Multi-objective Optimization by Learning Space Partition
Yiyang Zhao, Linnan Wang, Kevin Yang, Tianjun Zhang, Tian Guo, Yuandong Tian
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 |
|---|---|---|---|
| Add learned space partitioning to qEHVI | qEHVI · Nasbench201 | sample efficiency: 225% | Improved |
| Add learned space partitioning to CMA-ES | CMA-ES · BraninCurrin | samples to reach same hypervolume: 62.5% as many samples | Improved |
| Add learned space partitioning to CMA-ES | CMA-ES · VehicleSafety | samples to reach same hypervolume: 8% | Improved |
| Add learned space partitioning to CMA-ES | CMA-ES · Nasbench201 | samples to reach same hypervolume: 29% | Improved |
| Add learned space partitioning to qEHVI | qEHVI · BraninCurrin, VehicleSafety, and Nasbench201 | samples to achieve same performance: 50% samples | Improved |
| Add learned space partitioning to qEHVI | qEHVI · Molecule discovery | samples to achieve same performance: 71% fewer samples | Improved |
| Add learned space partitioning to CMA-ES | CMA-ES · Molecule discovery | samples to achieve same performance: 31% fewer samples | Improved |
| Add learned space partitioning to qEHVI | qEHVI · Molecule discovery | hypervolume: up to 10% | Improved |
| Replace random sampling with qEHVI sampling (ablation) | LaMOO on VehicleSafety · VehicleSafety | hypervolume | Improved |
| Replace random sampling with CMA-ES sampling (ablation) | LaMOO on VehicleSafety · VehicleSafety | hypervolume | Improved |
| Set the exploration factor Cp to 0 (ablation) | LaMOO on VehicleSafety · VehicleSafety | hypervolume | Hurt |
| Over-explore using a larger Cp (ablation) | LaMOO on VehicleSafety · VehicleSafety | hypervolume | Hurt |
Closest papers by meaning
- Interactive Hyperparameter Optimization in Multi-Objective Problems via Preference Learning (AAAI 2024 · similarity 0.87)
- ParetoFlow: Guided Flows in Multi-Objective Optimization (ICLR 2025 · similarity 0.86)
- SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization (NeurIPS 2026 · similarity 0.86)
- Preference-Guided Diffusion for Multi-Objective Offline Optimization (NeurIPS 2025 · similarity 0.86)
- BOtied: Multi-objective Bayesian optimization with tied multivariate ranks (ICML 2024 · similarity 0.85)
- Learning Configurations for Data-Driven Multi-Objective Optimization (ICML 2025 · similarity 0.85)
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.
