mlsanityData to 14 Oct 2026

Towards Explaining Distribution Shifts

arXiv 2210.10275 ↗First posted 19 Oct 2022cs.LG cs.CVcited by 26

Not on the map. Why?

Abstract (arXiv)

A distribution shift can have fundamental consequences such as signaling a change in the operating environment or significantly reducing the accuracy of downstream models. Thus, understanding distribution shifts is critical for examining and hopefully mitigating the effect of such a shift. Most prior work focuses on merely detecting if a shift has occurred and assumes any detected shift can be understood and handled appropriately by a human operator. We hope to aid in these manual mitigation tasks by explaining the distribution shift using interpretable transportation maps from the original distribution to the shifted one. We derive our interpretable mappings from a relaxation of optimal transport, where the candidate mappings are restricted to a set of interpretable mappings. We then inspect multiple quintessential use-cases of distribution shift in real-world tabular, text, and image datasets to showcase how our explanatory mappings provide a better balance between detail and interpretability than baseline explanations by both visual inspection and our PercentExplained metric.

Key facts. Cited by 26 papers in the map (data through 14 Oct 2026). Builds on 24 of its 47 references that are in the map. First posted on arXiv 19 Oct 2022.

Builds on

24 of 47 references.

References · 24

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  4. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  6. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  7. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  8. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,375
  9. 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,039
  10. 2018Computational Optimal Transportcited by 780
  11. 2019Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classificationcited by 404
  12. 2018Detecting and Correcting for Label Shift with Black Box Predictorscited by 366
12 more
  1. 2018Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shiftcited by 243
  2. 2019Optimal transport mapping via input convex neural networkscited by 206
  3. 2019Learning Model-Agnostic Counterfactual Explanations for Tabular Datacited by 146
  4. 2021Image-to-Image Translation: Methods and Applicationscited by 116
  5. 2019DIVA: Domain Invariant Variational Autoencoderscited by 115
  6. 2021Counterfactual Generative Networkscited by 111
  7. 2019Wasserstein-2 Generative Networkscited by 97
  8. 2017Learning Disentangled Representations with Semi-Supervised Deep Generative Modelscited by 72
  9. 2021Why did the distribution change?cited by 37
  10. 2021A Survey on Optimal Transport for Machine Learning: Theory and Applicationscited by 35
  11. 2021Feature Shift Detection: Localizing Which Features Have Shifted via Conditional Distribution Testscited by 24
  12. 2021Unified Shapley Framework to Explain Prediction Driftcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 3; 2024: 10; 2025: 8; 2026: 5.

26 papers.

Sort:

By area:

  1. 2023Scalable Optimal Transport Methods in Machine Learning: A Contemporary Surveycited by 27
  2. 2024A Survey on Evaluation of Out-of-Distribution Generalizationcited by 21
  3. 2024Counterfactual Fairness by Combining Factual and Counterfactual Predictionscited by 8
  4. 2024Self-Healing Machine Learning: A Framework for Autonomous Adaptation in Real-World Environmentscited by 8
  5. 2024Distributional Counterfactual Explanations With Optimal Transportcited by 6
  6. 2024A hierarchical decomposition for explaining ML performance discrepanciescited by 5
  7. 2024Enhancing Uncertainty Quantification in Drug Discovery with Censored Regression Labelscited by 5
  8. 2025PINS: Proximal Iterations with Sparse Newton and Sinkhorn for Optimal Transportcited by 5
  9. 2025GraphChain: Large Language Models for Large-scale Graph Analysis via Tool Chainingcited by 5
  10. 2023Towards Characterizing Domain Counterfactuals For Invertible Latent Causal Modelscited by 4
  11. 2025Unsupervised Detection of Distribution Shift in Inverse Problems using Diffusion Modelscited by 4
  12. 2025Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settingscited by 4
14 more
  1. 2024Recent Advances, Applications, and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2023 Symposiumcited by 3
  2. 2024Multiply-Robust Causal Change Attributioncited by 3
  3. 2025Revealing the Intrinsic Ethical Vulnerability of Aligned Large Language Modelscited by 2
  4. 2025Wasserstein Distances Made Explainable: Insights Into Dataset Shifts and Transport Phenomenacited by 2
  5. 2023Rectifying Group Irregularities in Explanations for Distribution Shiftcited by 1
  6. 2025Bayesian-based Online Label Shift Estimation with Dynamic Dirichlet Priorscited by 1
  7. 2024Control+Shift: Generating Controllable Distribution Shiftscited by 0
  8. 2024From XAI to MLOps: Explainable Concept Drift Detection with Profile Drift Detectioncited by 0
  9. 2025Explaining Concept Shift with Interpretable Feature Attributioncited by 0
  10. 2026ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Valuescited by 0
  11. 2026LatentDiff: Scaling Semantic Dataset Comparison to Millions of Imagescited by 0
  12. 2026Fitting Unknown Number of Hyperplanes with Manifold Optimizationcited by 0
  13. 2026Train, Test, Re-evaluate: Schedule-Sensitive Evaluation of Generative Data for Hand Detectioncited by 0
  14. 2026Curvature-Aware Zeroth-Order Optimization for Memory-Efficient Test-Time Adaptationcited by 0