Research map / Probabilistic and scientific modeling / Bayesian optimization
Multi-objective Bayesian optimization
Approximate Pareto fronts with batch, diversity, and preference-aware acquisition.
33 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Bayesian optimization, keyword cluster “acquisition · bayesian · multi objective”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
7 papers in 2023–24 and 14 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 7.6% to 10.4%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Predictive Entropy Search for Multi-objective Bayesian Optimization (ICML 2016 · 143 citations)
- 2018 · Batch Bayesian Optimization via Multi-objective Acquisition Ensemble for Automated Analog Circuit Design (ICML 2018 · 80 citations)
- 2019 · Max-value Entropy Search for Multi-Objective Bayesian Optimization (NeurIPS 2019 · 72 citations)
- 2020 · Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization (NeurIPS 2020 · 121 citations)
- 2021 · Parallel Bayesian Optimization of Multiple Noisy Objectives with Expected Hypervolume Improvement (NeurIPS 2021 · 40 citations)
- 2022 · Joint Entropy Search for Multi-Objective Bayesian Optimization (NeurIPS 2022 · 13 citations)
- 2023 · Sample-efficient Multi-objective Molecular Optimization with GFlowNets (NeurIPS 2023 · 6 citations)
- 2024 · Are You Concerned about Limited Function Evaluations: Data-Augmented Pareto Set Learning for Expensive Multi-Objective Optimization (AAAI 2024 · 11 citations)
- 2025 · Expensive Multi-Objective Bayesian Optimization Based on Diffusion Models (AAAI 2025 · 3 citations)
- 2026 · SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization (NeurIPS 2026 · 0 citations)
Earliest papers here
- Predictive Entropy Search for Multi-objective Bayesian Optimization (ICML 2016 · 143 citations)
- Pareto Frontier Learning with Expensive Correlated Objectives (ICML 2016 · 32 citations)
- Batch Bayesian Optimization via Multi-objective Acquisition Ensemble for Automated Analog Circuit Design (ICML 2018 · 80 citations)
What has been tried
25 papers here report 208 measured changes (a component added, removed or swapped, compared with a baseline): helped 64%, no clear effect 10%, hurt 20%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Remove data augmentation | Synthetic benchmarks and real-world applications · Pareto front convergence and diversity | Hurt | Are You Concerned about Limited Function Evaluations: Data-Augmented Pareto Set Learning for Expensive Multi-Objective Optimization (AAAI 2024 · 11 citations) | |
| Use adaptive acquisition-function selection instead of a static acquisition function | MOO benchmarks, batch size 1 · performance | Helped | Pareto Front-Diverse Batch Multi-Objective Bayesian Optimization (AAAI 2024 · 9 citations) | |
| Use both pairwise comparisons and improvement requests instead of only pairwise comparisons | DTLZ3 · simple regret | Helped | Multi-Objective Bayesian Optimization with Active Preference Learning (AAAI 2024 · 8 citations) | |
| add entropy-weighted gradients instead of mean-weighted gradients | ZDT1, ZDT2, ZDT3, DTLZ3, DTLZ6, DTLZ7 · HV | 4.9542e+00 -> 5.6570e+00; 5.4062e+00 -> 6.1533e+00; 4.2659e+00 -> 5.1703e+00; 1.0391e+10 -> 1.0795e+10; 4.3509e+03 -> 4. | Helped | Expensive Multi-Objective Bayesian Optimization Based on Diffusion Models (AAAI 2025 · 3 citations) |
| Add the local kernel | RE37, RE41, and RE42 · LHD | Helped | Improving Pareto Set Learning for Expensive Multi-objective Optimization via Stein Variational Hypernetworks (AAAI 2025 · 3 citations) | |
| Add local models to MLPS | QED+SA+GSK3β+JNK3 · HV | Helped | Multi-Objective Molecular Design Through Learning Latent Pareto Set (AAAI 2025 · 3 citations) | |
| Fit the CDF on pooled posterior samples rather than posterior means | Penicillin (d=7, M=3, q=1) · I_CDF | 0.20 (0.02) -> 0.26 (0.01) | Hurt | BOtied: Multi-objective Bayesian optimization with tied multivariate ranks (ICML 2024 · 1 citation) |
| Use MOCOBO to find K=2 solutions instead of using T individually optimized solutions | Rover, T=3 pairwise-conflicting courses, K=2 · average coverage score | 2.932 vs 12.899 | Hurt | Covering Multiple Objectives with a Small Set of Solutions Using Bayesian Optimization (NeurIPS 2025 · 0 citations) |
| Add a monotonicity constraint | various test problems · simple regret | Helped | Bayesian Optimization with Preference Exploration using a Monotonic Neural Network Ensemble (NeurIPS 2025 · 0 citations) | |
| Use a dynamically updated quality-uncertainty Pareto front when selecting batch solutions | 14 synthetic and practical problems · regret | Helped | Trading Off Quality and Uncertainty Through Multi-Objective Optimisation in Batch Bayesian Optimisation (AAAI 2025) |
What reviewers will expect
Datasets and baselines named in the 10 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- ZDT1 · 3 of 10 papers
- ZDT2 · 3 of 10 papers
- DTLZ2 · 3 of 10 papers
- ZDT3 · 2 of 10 papers
- DTLZ3 · 2 of 10 papers
- ZDT4 · 2 of 10 papers
- ZDT6 · 2 of 10 papers
Baselines
- qNEHVI · 3 of 10 papers
- MOEA/D-EGO · 2 of 10 papers
- TSEMO · 2 of 10 papers
- USeMO-EI · 2 of 10 papers
- DGEMO · 2 of 10 papers
- PSL-MOBO · 2 of 10 papers
- qNParEGO · 2 of 10 papers
- qEHVI · 2 of 10 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 21 papers; from OpenAlex and the papers' affiliations).
- East China Normal University · 2 papers
- Washington State University · 2 papers
- Stanford University · 2 papers
- Southern University of Science and Technology · 2 papers
Use it
Download BibTeX of 32 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Acquisition function design (97 papers)
- Local and high-dimensional Bayesian optimization (90 papers)
- Multi-fidelity hyperparameter optimization (44 papers)
- Regret bounds for Bayesian optimization (43 papers)
Elsewhere in the topic
- Offline black-box optimization with surrogates (56 papers)
- High-dimensional Bayesian optimization (34 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).