mlsanityData to 7 Oct 2026

Model Transferability With Responsive Decision Subjects

arXiv 2107.05911 ↗First posted 13 Jul 2021cs.LG stat.MLcited by 9

Not on the map. Why?

Abstract (arXiv)

Given an algorithmic predictor that is accurate on some source population consisting of strategic human decision subjects, will it remain accurate if the population respond to it? In our setting, an agent or a user corresponds to a sample (X,Y) drawn from a distribution calD and will face a model h and its classification result h(X). Agents can modify X to adapt to h, which will incur a distribution shift on (X,Y). Our formulation is motivated by applications where the deployed machine learning models are subjected to human agents, and will ultimately face responsive and interactive data distributions. We formalize the discussions of the transferability of a model by studying how the performance of the model trained on the available source distribution (data) would translate to the performance on its induced domain. We provide both upper bounds for the performance gap due to the induced domain shift, as well as lower bounds for the trade-offs that a classifier has to suffer on either the source training distribution or the induced target distribution. We provide further instantiated analysis for two popular domain adaptation settings, including covariate shift and target shift.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 29 of its 67 references that are in the map. First posted on arXiv 13 Jul 2021.

Builds on

29 of 67 references.

References · 29

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  2. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 762
  3. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 762
  4. 2013Domain Generalization via Invariant Feature Representationcited by 622
  5. 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 439
  6. 2019Contrastive Adaptation Network for Unsupervised Domain Adaptationcited by 407
  7. 2018Detecting and Correcting for Label Shift with Black Box Predictorscited by 364
  8. 2016European Union regulations on algorithmic decision-making and a "right to explanation"cited by 343
  9. 2016Unsupervised Domain Adaptation with Residual Transfer Networkscited by 340
  10. 2019Bridging Theory and Algorithm for Domain Adaptationcited by 332
  11. 2018Actionable Recourse in Linear Classificationcited by 327
  12. 2020Performative Predictioncited by 318
17 more
  1. 2018Adversarial Attacks and Defences: A Surveycited by 259
  2. 2015Strategic Classificationcited by 256
  3. 2020Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shiftcited by 235
  4. 2017Strategic Classification from Revealed Preferencescited by 126
  5. 2019Regularized Learning for Domain Adaptation under Label Shiftscited by 118
  6. 2019Strategic Classification is Causal Modeling in Disguisecited by 92
  7. 2020Stochastic Optimization for Performative Predictioncited by 92
  8. 2020Performative Prediction in a Stateful Worldcited by 90
  9. 2018How Do Classifiers Induce Agents To Invest Effort Strategically?cited by 82
  10. 2021How to Learn when Data Reacts to Your Model: Performative Gradient Descentcited by 70
  11. 2020Causal Strategic Linear Regressioncited by 65
  12. 2020Stochastic optimization with decision-dependent distributionscited by 65
  13. 2018Improving the Generalization of Adversarial Training with Domain Adaptationcited by 48
  14. 2020Maximizing Welfare with Incentive-Aware Evaluation Mechanismscited by 46
  15. 2020Test-time Unsupervised Domain Adaptationcited by 37
  16. 2020Domain Adaptation as a Problem of Inference on Graphical Modelscited by 36
  17. 2021Zeroth-Order Methods for Convex-Concave Minmax Problems: Applications to Decision-Dependent Risk Minimizationcited by 18

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 5; 2024: 1; 2025: 1.

9 papers.

  1. 2023Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignmentcited by 394
  2. 2022Fairness Transferability Subject to Bounded Distribution Shiftcited by 28
  3. 2022What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspectivecited by 18
  4. 2023Performative Prediction with Bandit Feedback: Learning through Reparameterizationcited by 11
  5. 2023The Importance of Human-Labeled Data in the Era of LLMscited by 8
  6. 2023Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factorscited by 7
  7. 2023Procedural Fairness Through Decoupling Objectionable Data Generating Componentscited by 3
  8. 2024Evaluating and Correcting Performative Effects of Decision Support Systems via Causal Domain Shiftcited by 3
  9. 2025To Give or Not to Give? The Impacts of Strategically Withheld Recoursecited by 2