mlsanityData to 7 Oct 2026

Decision-Focused Learning: Foundations, State of the Art, Benchmark and Future Opportunities

arXiv 2307.13565 ↗First posted 25 Jul 2023cs.LG cs.AI math.OCcited by 129

Not on the map. Why?

Abstract (arXiv)

Decision-focused learning (DFL) is an emerging paradigm that integrates machine learning (ML) and constrained optimization to enhance decision quality by training ML models in an end-to-end system. This approach shows significant potential to revolutionize combinatorial decision-making in real-world applications that operate under uncertainty, where estimating unknown parameters within decision models is a major challenge. This paper presents a comprehensive review of DFL, providing an in-depth analysis of both gradient-based and gradient-free techniques used to combine ML and constrained optimization. It evaluates the strengths and limitations of these techniques and includes an extensive empirical evaluation of eleven methods across seven problems. The survey also offers insights into recent advancements and future research directions in DFL. Code and benchmark: https://github.com/PredOpt/predopt-benchmarks

Key facts. Cited by 129 papers in the map (data through 7 Oct 2026). Builds on 58 of its 130 references that are in the map. First posted on arXiv 25 Jul 2023.

Builds on

58 of 130 references.

References · 58

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  5. 2015Automatic differentiation in machine learning: a surveycited by 846
  6. 2017OptNet: Differentiable Optimization as a Layer in Neural Networkscited by 588
  7. 2019Differentiable Convex Optimization Layerscited by 458
  8. 2019Algorithm Unrolling: Interpretable, Efficient Deep Learning for Signal and Image Processingcited by 411
  9. 2016From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classificationcited by 337
  10. 2017Smart "Predict, then Optimize"cited by 332
  11. 2019Differentiation of Blackbox Combinatorial Solverscited by 224
  12. 2018Melding the Data-Decisions Pipeline: Decision-Focused Learning for Combinatorial Optimizationcited by 220
38 more
  1. 2020Fast Differentiable Sorting and Rankingcited by 196
  2. 2018Fairness of Exposure in Rankingscited by 189
  3. 2019SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solvercited by 173
  4. 2014From Predictive to Prescriptive Analyticscited by 142
  5. 2019Learning Dense Representations for Entity Retrievalcited by 122
  6. 2016On Differentiating Parameterized Argmin and Argmax Problems with Application to Bi-level Optimizationcited by 108
  7. 2021Implicit MLE: Backpropagating Through Discrete Exponential Family Distributionscited by 97
  8. 2019MIPaaL: Mixed Integer Program as a Layercited by 93
  9. 2019Smart Predict-and-Optimize for Hard Combinatorial Optimization Problemscited by 88
  10. 2020Interior Point Solving for LP-based prediction+optimisationcited by 83
  11. 2020Learning with Differentiable Perturbed Optimizerscited by 81
  12. 2020Deep Graph Matching via Blackbox Differentiation of Combinatorial Solverscited by 81
  13. 2023A Survey of Contextual Optimization Methods for Decision Making under Uncertaintycited by 81
  14. 2019Differentiating Through a Cone Programcited by 72
  15. 2021Differentiable Patch Selection for Image Recognitioncited by 68
  16. 2020Gradient Estimation with Stochastic Softmax Trickscited by 66
  17. 2021Decision-Focused Learning: Through the Lens of Learning to Rankcited by 61
  18. 2019Optimizing Rank-based Metrics with Blackbox Differentiationcited by 58
  19. 2019Generalization Bounds in the Predict-then-Optimize Frameworkcited by 54
  20. 2020Decision Trees for Decision-Making under the Predict-then-Optimize Frameworkcited by 52
  21. 2020Contrastive Losses and Solution Caching for Predict-and-Optimizecited by 52
  22. 2021CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraintscited by 49
  23. 2019End to end learning and optimization on graphscited by 44
  24. 2022PyEPO: A PyTorch-based End-to-End Predict-then-Optimize Library for Linear and Integer Programmingcited by 32
  25. 2022Backpropagation through Combinatorial Algorithms: Identity with Projection Workscited by 29
  26. 2021Risk Bounds and Calibration for a Smart Predict-then-Optimize Methodcited by 28
  27. 2022Learning with Combinatorial Optimization Layers: a Probabilistic Approachcited by 28
  28. 2018Solution Refinement at Regular Points of Conic Problemscited by 21
  29. 2019The Limited Multi-Label Projection Layercited by 21
  30. 2022SurCo: Learning Linear Surrogates For Combinatorial Nonlinear Optimization Problemscited by 21
  31. 2014On distinguishability criteria for estimating generative modelscited by 20
  32. 2020Learning Linear Programs from Optimal Decisionscited by 18
  33. 2021End-to-end Learning for Fair Ranking Systemscited by 18
  34. 2021The Perils of Learning Before Optimizingcited by 15
  35. 2023Backpropagation of Unrolled Solvers with Folded Optimizationcited by 13
  36. 2022Predict+Optimize for Packing and Covering LPs with Unknown Parameters in Constraintscited by 11
  37. 2023Electricity Price Prediction for Energy Storage System Arbitrage: A Decision-focused Approachcited by 7
  38. 2023Active Learning For Contextual Linear Optimization: A Margin-Based Approachcited by 6

50 most cited of 58 All 58 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 6; 2024: 19; 2025: 41; 2026: 63.

129 papers.

Sort:

By area:

  1. 2025Training-Free Constrained Generation With Stable Diffusion Modelscited by 20
  2. 2025FSNet: Feasibility-Seeking Neural Network for Constrained Optimization with Guaranteescited by 19
  3. 2025Gen-DFL: Decision-Focused Generative Learning for Robust Decision Makingcited by 18
  4. 2024End-to-End Conformal Calibration for Optimization Under Uncertaintycited by 14
  5. 2025Decision-informed Neural Networks with Large Language Model Integration for Portfolio Optimizationcited by 14
  6. 2024Offline Model-Based Optimization by Learning to Rankcited by 13
  7. 2025Conformal Risk Training: End-to-End Optimization of Conformal Risk Controlcited by 11
  8. 2024Decision Focused Causal Learning for Direct Counterfactual Marketing Optimizationcited by 9
  9. 2023CaVE: A Cone-Aligned Approach for Fast Predict-then-optimize with Binary Linear Programscited by 8
  10. 2024Differentiable Distributionally Robust Optimization Layerscited by 8
  11. 2025Solver-Free Decision-Focused Learning for Linear Optimization Problemscited by 8
  12. 2024Towards a Systems Theory of Algorithmscited by 7
38 more
  1. 2024Return Prediction for Mean-Variance Portfolio Selection: How Decision-Focused Learning Shapes Forecasting Modelscited by 5
  2. 2025Decision-Focused Fine-Tuning of Time Series Foundation Models for Dispatchable Feeder Optimizationcited by 5
  3. 2025Feasibility-Aware Decision-Focused Learning for Predicting Parameters in the Constraintscited by 5
  4. 2025Diffusion-DFL: Decision-focused Diffusion Models for Stochastic Optimizationcited by 5
  5. 2023Predict-Then-Optimize by Proxy: Learning Joint Models of Prediction and Optimizationcited by 4
  6. 2024Uplift modeling with continuous treatments: A predict-then-optimize approachcited by 4
  7. 2025Global-Decision-Focused Neural ODEs for Proactive Grid Resilience Managementcited by 4
  8. 2025Online Decision-Focused Learningcited by 4
  9. 2025Calibrating Decision Robustness via Inverse Conformal Risk Controlcited by 4
  10. 2023Learning Lagrangian Multipliers for the Travelling Salesman Problemcited by 3
  11. 2024Learning Joint Models of Prediction and Optimizationcited by 3
  12. 2024Self-Supervised Learning of Iterative Solvers for Constrained Optimizationcited by 3
  13. 2025Conformalized Decision Risk Assessmentcited by 3
  14. 2025A Fully First-Order Layer for Differentiable Optimizationcited by 3
  15. 2026Combinatorial Optimization Augmented Machine Learningcited by 3
  16. 2026Decision-Focused Learning via Tangent-Space Projection of Prediction Errorcited by 3
  17. 2026Constraint-Aware Flow Matching: Decision Aligned End-to-End Training for Constrained Samplingcited by 3
  18. 2024End-to-End Learning for Fair Multiobjective Optimization Under Uncertaintycited by 2
  19. 2024Metalearners for Ranking Treatment Effectscited by 2
  20. 2024Decision-Focused Forecasting: A Differentiable Multistage Optimisation Architecturecited by 2
  21. 2025Welfare as a Guiding Principle for Machine Learning -- From Compass, to Lens, to Roadmapcited by 2
  22. 2025Sufficient Decision Proxies for Decision-Focused Learningcited by 2
  23. 2025Learning with Local Search MCMC Layerscited by 2
  24. 2025Smart Surrogate Losses for Contextual Stochastic Linear Optimization with Robust Constraintscited by 2
  25. 2025Estimating Covariance for Global Minimum Variance Portfolio: A Decision-Focused Learning Approachcited by 2
  26. 2025Minimizing Surrogate Losses for Decision-Focused Learning using Differentiable Optimizationcited by 2
  27. 2025Bi-Level Decision-Focused Causal Learning for Large-Scale Marketing Optimization: Bridging Observational and Experimental Datacited by 2
  28. 2025A Dual Perspective on Decision-Focused Learning: Scalable Training via Dual-Guided Surrogatescited by 2
  29. 2026Decision-Focused Sequential Experimental Design: A Directional Uncertainty-Guided Approachcited by 2
  30. 2026Decision-calibrated prediction sets for robust power system operationscited by 2
  31. 2026A Solver-Free Training Method for Predict-then-Optimizecited by 2
  32. 2024Learning Fair Ranking Policies via Differentiable Optimization of Ordered Weighted Averagescited by 1
  33. 2024Differentiation of Multi-objective Data-driven Decision Pipelinecited by 1
  34. 2024DistrictNet: Decision-aware learning for geographical districtingcited by 1
  35. 2025Decision-aware training of spatiotemporal forecasting models to select a top K subset of sites for interventioncited by 1
  36. 2025Decision by Supervised Learning with Deep Ensembles: A Practical Framework for Robust Portfolio Optimizationcited by 1
  37. 2025Aligning the Evaluation of Probabilistic Predictions with Downstream Valuecited by 1
  38. 2025Decision-Focused Learning Enhanced by Automated Feature Engineering for Energy Storage Optimisationcited by 1

50 most cited of 129 All 129 (JSON)