mlsanityData to 7 Oct 2026

A Near-Optimal Algorithm for Debiasing Trained Machine Learning Models

arXiv 2106.12887 ↗First posted 6 Jun 2021cs.LG cs.AI stat.MLcited by 19

Not on the map. Why?

Abstract (arXiv)

We present a scalable post-processing algorithm for debiasing trained models, including deep neural networks (DNNs), which we prove to be near-optimal by bounding its excess Bayes risk. We empirically validate its advantages on standard benchmark datasets across both classical algorithms as well as modern DNN architectures and demonstrate that it outperforms previous post-processing methods while performing on par with in-processing. In addition, we show that the proposed algorithm is particularly effective for models trained at scale where post-processing is a natural and practical choice.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 30 of its 70 references that are in the map. First posted on arXiv 6 Jun 2021.

Builds on

30 of 70 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2016Densely Connected Convolutional Networkscited by 6,691
  4. 2017On Calibration of Modern Neural Networkscited by 4,860
  5. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  6. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  7. 2021Convex optimizationcited by 2,479
  8. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  9. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  10. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  11. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
  12. 2017Revisiting Unreasonable Effectiveness of Data in Deep Learning Eracited by 916
18 more
  1. 2019Big Transfer (BiT): General Visual Representation Learningcited by 766
  2. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 671
  3. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  4. 2018A Reductions Approach to Fair Classificationcited by 585
  5. 2016Inherent Trade-Offs in the Fair Determination of Risk Scorescited by 582
  6. 2017Algorithmic decision making and the cost of fairnesscited by 402
  7. 2016Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatmentcited by 368
  8. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  9. 2019Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigationcited by 252
  10. 2019Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchycited by 156
  11. 2019Does Object Recognition Work for Everyone?cited by 149
  12. 2018Classification with Fairness Constraints: A Meta-Algorithm with Provable Guaranteescited by 146
  13. 2019On the Fairness of Disentangled Representationscited by 144
  14. 2020REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasetscited by 136
  15. 2020On Robustness and Transferability of Convolutional Neural Networkscited by 98
  16. 2016A Confidence-Based Approach for Balancing Fairness and Accuracycited by 73
  17. 2019Recovering from Biased Data: Can Fairness Constraints Improve Accuracy?cited by 50
  18. 2016A statistical framework for fair predictive algorithmscited by 37

Built on by

Cited by year

Citing papers per year: 2022: 11; 2023: 2; 2024: 5; 2025: 1.

19 papers.

Sort:

  1. 2023Scaling Vision Transformers to 22 Billion Parameterscited by 690
  2. 2022Extremely Simple Activation Shaping for Out-of-Distribution Detectioncited by 212
  3. 2022Bias Mitigation for Machine Learning Classifiers: A Comprehensive Surveycited by 125
  4. 2022Detecting Shortcut Learning for Fair Medical AI using Shortcut Testingcited by 52
  5. 2022Diagnosing failures of fairness transfer across distribution shift in real-world medical settingscited by 45
  6. 2022On Data Scaling in Masked Image Modelingcited by 45
  7. 2024CLIP the Bias: How Useful is Balancing Data in Multimodal Learning?cited by 35
  8. 2022Multicalibrated Regression for Downstream Fairnesscited by 13
  9. 2022Are Two Heads the Same as One? Identifying Disparate Treatment in Fair Neural Networkscited by 10
  10. 2022A Reduction to Binary Approach for Debiasing Multiclass Datasetscited by 9
  11. 2024Mind the Graph When Balancing Data for Fairness or Robustnesscited by 9
  12. 2023Fairness-aware Vision Transformer via Debiased Self-Attentioncited by 6
7 more
  1. 2022Fair Wrapping for Black-box Predictionscited by 4
  2. 2022Fairness via In-Processing in the Over-parameterized Regime: A Cautionary Talecited by 4
  3. 2024Looking Beyond What You See: An Empirical Analysis on Subgroup Intersectional Fairness for Multi-label Chest X-ray Classification Using Social Determinants of Racial Health Inequitiescited by 3
  4. 2022Layer-Stack Temperature Scalingcited by 2
  5. 2024Achievable Fairness on Your Data With Utility Guaranteescited by 2
  6. 2024Personalised Drug Identifier for Cancer Treatment with Transformers using Auxiliary Informationcited by 0
  7. 2025A Critical Review of Predominant Bias in Neural Networkscited by 0