Research map / Learning theory and algorithms / Combinatorial and search optimization
Decision-focused optimization
Integrates prediction with optimization, including data requirements, robustness and dynamic programming for budgeted search.
97 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Combinatorial and search optimization, keyword cluster “objective · multi · solutions”
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.
22 papers in 2023–24 and 23 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 6.3% to 6.3%.
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 · Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach (NeurIPS 2016 · 56 citations)
- 2017 · Efficient Parameter Importance Analysis via Ablation with Surrogates (AAAI 2017 · 36 citations)
- 2018 · Resource-Constrained Scheduling for Maritime Traffic Management (AAAI 2018 · 15 citations)
- 2019 · Improving Optimization Bounds Using Machine Learning: Decision Diagrams Meet Deep Reinforcement Learning (AAAI 2019 · 43 citations)
- 2020 · Smart Predict-and-Optimize for Hard Combinatorial Optimization Problems (AAAI 2020 · 108 citations)
- 2021 · Escaping Local Optima with Non-Elitist Evolutionary Algorithms (AAAI 2021 · 35 citations)
- 2022 · Sample Efficiency Matters: A Benchmark for Practical Molecular Optimization (NeurIPS 2022 · 57 citations)
- 2023 · Slimmable Dataset Condensation (CVPR 2023 · 40 citations)
- 2024 · E2E-AT: A Unified Framework for Tackling Uncertainty in Task-Aware End-to-End Learning (AAAI 2024 · 6 citations)
- 2025 · Fast Contiguous Somatic Hypermutations for Single-Objective Optimisation and Multi-Objective Optimisation Via Decomposition (AAAI 2025 · 3 citations)
- 2026 · Soft Quality-Diversity Optimization (ICLR 2026 · 0 citations)
Earliest papers here
- Bayesian Optimization with a Finite Budget: An Approximate Dynamic Programming Approach (NeurIPS 2016 · 56 citations)
- Multistage Campaigning in Social Networks (NeurIPS 2016 · 30 citations)
- Linear Relaxations for Finding Diverse Elements in Metric Spaces (NeurIPS 2016 · 15 citations)
What has been tried
81 papers here report 648 measured changes (a component added, removed or swapped, compared with a baseline): helped 64%, no clear effect 8%, hurt 20%, mixed 8%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Set α=0.5 rather than α=1.0 in AT-INPUT with εx=0.02 | NCED, integrated attack (0.02, 0.05) · task-aware cost | 934.4 -> 914.8 | Helped | E2E-AT: A Unified Framework for Tackling Uncertainty in Task-Aware End-to-End Learning (AAAI 2024 · 6 citations) |
| add entropy loss | All datasets (average) · R^2 | 0.9743 -> 0.9859 | Helped | MetaSymNet: A Tree-like Symbol Network with Adaptive Architecture and Activation Functions (AAAI 2025 · 2 citations) |
| Set the number of clusters k below the maximum width W | KP · bound quality | No clear effect | Using Clustering to Strengthen Decision Diagram Bounds for Discrete Optimization (AAAI 2024 · 2 citations) | |
| remove adaptive risk management and use normal gradient ascent | Welded beam design · performance | Hurt | Constrained Offline Black-Box Optimization via Risk Evaluation and Management (AAAI 2025 · 1 citation) | |
| Increase the number of participating individuals | Galápagos tortoises, 45–51 individuals · mean pairwise relatedness | Mixed | Constraint Optimisation Approaches for Designing Group-Living Captive Breeding Programmes (AAAI 2025 · 1 citation) | |
| Use the pure bound rather than the Lex or Weighted side-constraint-aware bounds | Mandatory-vertex instances · proof construction success rate | 96.29% vs. 93.57%; 96.29% vs. 95.93% | Helped | Paths, Proofs, and Perfection: Developing a Human-Interpretable Proof System for Constrained Shortest Paths (AAAI 2024 · 1 citation) |
| Set ε to 0.1 rather than 0 | Random LP · normalized regret | 0.023 ± 0.001 -> 0.016 ± 0.001 | Helped | Solver-Free Decision-Focused Learning for Linear Optimization Problems (NeurIPS 2025 · 0 citations) |
| Measure solution similarity only across the three bridge edges instead of all edges | Synthetic shortest path · optimality reduction for fully explainable solutions | 10% | Hurt | A Framework for Data-Driven Explainability in Mathematical Optimization (AAAI 2024) |
What reviewers will expect
Datasets and baselines named in the 15 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- MSE · 2 of 15 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 45 papers; from OpenAlex and the papers' affiliations).
- National University of Singapore · 4 papers
- Massachusetts Institute of Technology · 3 papers
- University of Virginia · 2 papers
- ETHZ - ETH Zurich · 2 papers
- KU Leuven · 2 papers
Use it
Download BibTeX of 40 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
- Constrained neural network optimization (111 papers)
- Unsupervised neural combinatorial optimization (113 papers)
- Multi-objective optimization (93 papers)
- Globally optimal geometric optimization (56 papers)
Elsewhere in the topic
- Logic and answer set programming theory (150 papers)
- Heuristic and local search algorithms (147 papers)
- Classical and HTN planning (145 papers)
- Structure and complexity of constraint reasoning (92 papers)
- Predict-then-optimize learning (89 papers)
- SAT and model counting solvers (89 papers)
- Heuristics for integer programming (60 papers)
- Differentiable and symbolic reasoning (56 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).