Research map / Probabilistic and scientific modeling / Bayesian optimization
Multi-Objective Bayesian Optimization via Adaptive $\varepsilon$-Constraint Decomposition
Yaohong Yang, Sammie Katt, Samuel Kaski
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 Thompson-sampled path instead of posterior mean for target identification (ablation) | STAGE-BO · unconstrained MOO (6 benchmarks) | log HV difference, IGD, IGD+ | Improved |
| use Thompson-sampled path instead of posterior mean (ablation) | STAGE-BO · Water Planning | IGD | No clear effect |
| add clipping of epsilon constraints to best observed value (ablation) | STAGE-BO · unconstrained MOO, most benchmarks | HV, IGD | No clear effect |
| add clipping of epsilon constraints (ablation) | STAGE-BO · Rocket Injector | HV, IGD | Improved |
| add clipping of epsilon constraints (ablation) | STAGE-BO · Gear Train (constrained) | HV, IGD | Improved |
| add clipping of epsilon constraints (ablation) | STAGE-BO · preference-based MOO | HV, IGD, IGD+ | No clear effect |
| add cEI acquisition instead of directly querying the maxmin target (ablation) | STAGE-BO · ZDT1 | convergence and coverage | Improved |
| use maxmin gap-filling targets instead of random constraints (ablation) | STAGE-BO · ZDT1 | convergence and coverage | Improved |
| replace round-robin objective selection with random or feasible-based selection (ablation) | STAGE-BO · ZDT1 | HV, IGD | No clear effect |
| use fill-distance-based epsilon constraints instead of hypervolume maximization | STAGE-BO vs qEHVI · ZDT1 and ZDT2 | IGD | Improved |
| use STAGE-BO versus baselines on constrained MOO | STAGE-BO · constrained MOO (MW7, CONSTR, Disc brake, Gear train) | IGD | Improved |
| use STAGE-BO versus baselines on Gear Train and Disc Brake | STAGE-BO · Gear Train, Disc Brake | IGD vs HV | Mixed |
Closest papers by meaning
- MOBO-OSD: Batch Multi-Objective Bayesian Optimization via Orthogonal Search Directions (NeurIPS 2025 · similarity 0.90)
- Preference-Aware Constrained Multi-Objective Bayesian Optimization (Student Abstract) (AAAI 2024 · similarity 0.90)
- BayeSQP: Bayesian Optimization through Sequential Quadratic Programming (NeurIPS 2025 · similarity 0.88)
- BOtied: Multi-objective Bayesian optimization with tied multivariate ranks (ICML 2024 · similarity 0.88)
- Pareto Front-Diverse Batch Multi-Objective Bayesian Optimization (AAAI 2024 · similarity 0.88)
- User Preference Meets Pareto-Optimality in Multi-Objective Bayesian Optimization (AAAI 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.
