mlsanityData to 7 Oct 2026

Optimising Equal Opportunity Fairness in Model Training

arXiv 2205.02393 ↗First posted 5 May 2022cs.LG cs.CLcited by 22

Not on the map. Why?

Abstract (arXiv)

Real-world datasets often encode stereotypes and societal biases. Such biases can be implicitly captured by trained models, leading to biased predictions and exacerbating existing societal preconceptions. Existing debiasing methods, such as adversarial training and removing protected information from representations, have been shown to reduce bias. However, a disconnect between fairness criteria and training objectives makes it difficult to reason theoretically about the effectiveness of different techniques. In this work, we propose two novel training objectives which directly optimise for the widely-used criterion of it equal opportunity, and show that they are effective in reducing bias while maintaining high performance over two classification tasks.

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Builds on 34 of its 55 references that are in the map. First posted on arXiv 5 May 2022.

Builds on

34 of 55 references.

References · 34

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  4. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  5. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  6. 2014Certifying and removing disparate impactcited by 715
  7. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  8. 2018A Reductions Approach to Fair Classificationcited by 585
  9. 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 442
  10. 2020Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamicscited by 423
  11. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  12. 2017On Fairness and Calibrationcited by 406
22 more
  1. 2016Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatmentcited by 368
  2. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  3. 2019Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Settingcited by 346
  4. 2020Fairness without Demographics through Adversarially Reweighted Learningcited by 251
  5. 2018Empirical Risk Minimization under Fairness Constraintscited by 226
  6. 2016Demographic Dialectal Variation in Social Media: A Case Study of African-American Englishcited by 194
  7. 2019Dice Loss for Data-imbalanced NLP Taskscited by 180
  8. 2018Adversarial Removal of Demographic Attributes from Text Datacited by 166
  9. 2018FairGAN: Fairness-aware Generative Adversarial Networkscited by 164
  10. 2017Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasmcited by 117
  11. 2020FairBatch: Batch Selection for Model Fairnesscited by 99
  12. 2019Conditional Learning of Fair Representationscited by 78
  13. 2018Towards Robust and Privacy-preserving Text Representationscited by 75
  14. 2019Training individually fair ML models with Sensitive Subspace Robustnesscited by 73
  15. 2020FR-Train: A Mutual Information-Based Approach to Fair and Robust Trainingcited by 52
  16. 2021Diverse Adversaries for Mitigating Bias in Trainingcited by 48
  17. 2019Average Individual Fairness: Algorithms, Generalization and Experimentscited by 46
  18. 2020Why resampling outperforms reweighting for correcting sampling bias with stochastic gradientscited by 34
  19. 2021Contrastive Learning for Fair Representationscited by 34
  20. 2021Evaluating Debiasing Techniques for Intersectional Biasescited by 30
  21. 2021Fairness-aware Class Imbalanced Learningcited by 19
  22. 2022fairlib: A Unified Framework for Assessing and Improving Classification Fairnesscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 7; 2024: 6; 2025: 4; 2026: 1.

22 papers.

Sort:

By area:

  1. 2022BLIND: Bias Removal With No Demographicscited by 22
  2. 2022Equal Improvability: A New Fairness Notion Considering the Long-term Impactcited by 13
  3. 2023Fair Enough: Standardizing Evaluation and Model Selection for Fairness Research in NLPcited by 11
  4. 2023Loss Balancing for Fair Supervised Learningcited by 11
  5. 2023Same Same, But Different: Conditional Multi-Task Learning for Demographic-Specific Toxicity Detectioncited by 10
  6. 2023iLab at SemEval-2023 Task 11 Le-Wi-Di: Modelling Disagreement or Modelling Perspectives?cited by 10
  7. 2023Fair Text Classification with Wasserstein Independencecited by 4
  8. 2024Addressing Both Statistical and Causal Gender Fairness in NLP Modelscited by 4
  9. 2022Systematic Evaluation of Predictive Fairnesscited by 3
  10. 2025Fair Text Classification via Transferable Representationscited by 3
  11. 2025Enhancing Multi-Attribute Fairness in Healthcare Predictive Modelingcited by 2
  12. 2022A Differentiable Distance Approximation for Fairer Image Classificationcited by 1
10 more
  1. 2023OccuQuest: Mitigating Occupational Bias for Inclusive Large Language Modelscited by 1
  2. 2023Selecting Shots for Demographic Fairness in Few-Shot Learning with Large Language Modelscited by 1
  3. 2024Configurable Fairness: Direct Optimization of Parity Metrics via Vision-Language Modelscited by 1
  4. 2024An ExplainableFair Framework for Prediction of Substance Use Disorder Treatment Completioncited by 1
  5. 2024Fairness-Aware Multi-Group Target Detection in Online Discussioncited by 1
  6. 2024Fair Class-Incremental Learning using Sample Weightingcited by 0
  7. 2024Mitigating Extrinsic Gender Bias for Bangla Classification Taskscited by 0
  8. 2025Benchmarking Gender and Political Bias in Large Language Modelscited by 0
  9. 2025Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysiscited by 0
  10. 2026When the Server Steps In: Calibrated Updates for Fair Federated Learningcited by 0