mlsanityData to 7 Oct 2026

Distributionally Robust Optimization via Generative Ambiguity Modeling

arXiv 2602.08976 ↗First posted 9 Feb 2026cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

This paper studies Distributionally Robust Optimization (DRO), a fundamental framework for enhancing the robustness and generalization of statistical learning and optimization. An effective ambiguity set for DRO must involve distributions that remain consistent to the nominal distribution while being diverse enough to account for a variety of potential scenarios. Moreover, it should lead to tractable DRO solutions. To this end, we propose generative model-based ambiguity sets that capture various adversarial distributions beyond the nominal support space while maintaining consistency with the nominal distribution. Building on this generative ambiguity modeling, we propose DRO with Generative Ambiguity Set (GAS-DRO), a tractable DRO algorithm that solves the inner maximization over the parameterized generative model space. We formally establish the stationary convergence performance of GAS-DRO. We implement GAS-DRO with a diffusion model and empirically demonstrate its superior Out-of-Distribution (OOD) generalization performance in ML tasks.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 42 of its 59 references that are in the map. First posted on arXiv 9 Feb 2026.

Builds on

42 of 59 references.

References · 42

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2020Denoising Diffusion Implicit Modelscited by 9,655
  4. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  5. 2019An Introduction to Variational Autoencoderscited by 997
  6. 2021Maximum Likelihood Training of Score-Based Diffusion Modelscited by 670
  7. 2021Towards Out-Of-Distribution Generalization: A Surveycited by 431
  8. 2022Understanding Diffusion Models: A Unified Perspectivecited by 337
  9. 2017Certifying Some Distributional Robustness with Principled Adversarial Trainingcited by 333
  10. 2023Your Diffusion Model is Secretly a Zero-Shot Classifiercited by 301
  11. 2015Data-driven Distributionally Robust Optimization Using the Wasserstein Metric: Performance Guarantees and Tractable Reformulationscited by 285
  12. 2023Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Spacecited by 183
30 more
  1. 2016Distributionally Robust Stochastic Optimization with Wasserstein Distancecited by 182
  2. 2019Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learningcited by 169
  3. 2018Learning to Learn from Noisy Labeled Datacited by 160
  4. 2017Learning From Noisy Large-Scale Datasets With Minimal Supervisioncited by 115
  5. 2023Robust Classification via a Single Diffusion Modelcited by 79
  6. 2019Distributionally Robust Optimization and Generalization in Kernel Methodscited by 73
  7. 2020Distributionally Robust Bayesian Optimizationcited by 60
  8. 2021Out of Distribution Generalization in Machine Learningcited by 59
  9. 2021Robot Learning from Randomized Simulations: A Reviewcited by 56
  10. 2023AdvDiff: Generating Unrestricted Adversarial Examples using Diffusion Modelscited by 46
  11. 2021Distributionally Robust Learningcited by 44
  12. 2019Distributionally Robust Reinforcement Learningcited by 40
  13. 2021Sinkhorn Distributionally Robust Optimizationcited by 38
  14. 2021Modeling the Second Player in Distributionally Robust Optimizationcited by 29
  15. 2024Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithmscited by 24
  16. 2025Gen-DFL: Decision-Focused Generative Learning for Robust Decision Makingcited by 18
  17. 2021Distributionally Robust Policy Learning via Adversarial Environment Generationcited by 17
  18. 2022Distributionally Robust Models with Parametric Likelihood Ratioscited by 17
  19. 2022Distributionally Robust Bayesian Optimization with varphi-divergencescited by 16
  20. 2023Flow-based Distributionally Robust Optimizationcited by 13
  21. 2023Unifying Distributionally Robust Optimization via Optimal Transport Theorycited by 12
  22. 2024Diffusion Models For Multi-Modal Generative Modelingcited by 9
  23. 2024Differentiable Distributionally Robust Optimization Layerscited by 8
  24. 2024Distributionally Robust Optimization via Iterative Algorithms in Continuous Probability Spacescited by 8
  25. 2021Adversarial Deep Learning for Online Resource Allocationcited by 7
  26. 2025Worst-case generation via minimax optimization in Wasserstein spacecited by 7
  27. 2020Conic Reformulations for Kullback-Leibler Divergence Constrained Distributionally Robust Optimization and Applicationscited by 4
  28. 2025Knowledge-Guided Wasserstein Distributionally Robust Optimizationcited by 3
  29. 2025Nested Stochastic Algorithm for Generalized Sinkhorn distance-Regularized Distributionally Robust Optimizationcited by 3
  30. 2025Generative Models for Synthetic Data: Transforming Data Mining in the GenAI Eracited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Generative models for decision-making under distributional shiftcited by 4
  2. 2026Sampler-Robust Optimization under Generative Modelscited by 0
  3. 2026Generative Distributionally Robust Optimizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Distributionally Robust Optimization via Diffusion Ambiguity Modeling
  2. 2025Improved Diffusion-based Generative Model with Better Adversarial Robustness
  3. 2023Achieving the Asymptotically Optimal Sample Complexity of Offline Reinforcement Learning: A DRO-Based Approach
  4. 2026Decision-Focused Scenario Generation and Selection for Efficient and Robust Grid Dispatch
  5. 2026PAPT++: Risk-Aware Adversarial Tuning and Generation for Single Domain Generalization
  6. 2024Score-based generative models are provably robust: an uncertainty quantification perspective
  7. 2025Generative Bayesian Optimization: Generative Models as Acquisition Functions
  8. 2026Learning Where to Steer: Noise-Space Geometry for Efficient Offline Multi-Objective Optimization with Generative Models