An Intersectional Definition of Fairness

@article{Foulds2020AnID,
  title={An Intersectional Definition of Fairness},
  author={James R. Foulds and Shimei Pan},
  journal={2020 IEEE 36th International Conference on Data Engineering (ICDE)},
  year={2020},
  pages={1918-1921}
}
  • James R. Foulds, Shimei Pan
  • Published 2020
  • Computer Science, Mathematics
  • 2020 IEEE 36th International Conference on Data Engineering (ICDE)
We propose differential fairness, a multi-attribute definition of fairness in machine learning which is informed by intersectionality, a critical lens arising from the humanities literature, leveraging connections between differential privacy and legal notions of fairness. We show that our criterion behaves sensibly for any subset of the set of protected attributes, and we prove economic, privacy, and generalization guarantees. We provide a learning algorithm which respects our differential… Expand
29 Citations
Fairness with Overlapping Groups
  • 2
  • PDF
Bayesian Modeling of Intersectional Fairness: The Variance of Bias
  • 7
  • PDF
Auditing and Achieving Intersectional Fairness in Classification Problems
  • 3
  • Highly Influenced
  • PDF
Transparency Tools for Fairness in AI (Luskin)
  • 1
  • PDF
Multi-Differential Fairness Auditor for Black Box Classifiers
  • 2
  • PDF
Causal Multi-Level Fairness
  • PDF
Causal intersectionality for fair ranking
  • 9
  • PDF
...
1
2
3
...

References

SHOWING 1-10 OF 61 REFERENCES
Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
  • 258
  • Highly Influential
  • PDF
A Convex Framework for Fair Regression
  • 117
  • PDF
Counterfactual Fairness
  • 463
  • Highly Influential
  • PDF
Fairness through awareness
  • 1,220
  • PDF
Empirical Risk Minimization under Fairness Constraints
  • 156
  • PDF
Interventional Fairness: Causal Database Repair for Algorithmic Fairness
  • 47
  • PDF
Differentially Private Fair Learning
  • 40
  • PDF
Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions
  • 79
  • PDF
Equality of Opportunity in Supervised Learning
  • 1,308
  • PDF
...
1
2
3
4
5
...