Conditional Learning of Fair Representations
arXiv:1910.07162
Abstract
We propose a novel algorithm for learning fair representations that can simultaneously mitigate two notions of disparity among different demographic subgroups in the classification setting. Two key components underpinning the design of our algorithm are balanced error rate and conditional alignment of representations. We show how these two components contribute to ensuring accuracy parity and equalized false-positive and false-negative rates across groups without impacting demographic parity. Furthermore, we also demonstrate both in theory and on two real-world experiments that the proposed algorithm leads to a better utility-fairness trade-off on balanced datasets compared with existing algorithms on learning fair representations for classification.
References in corpus (9)
- Equality of Opportunity in Supervised Learning
- Counterfactual Fairness
- The Measure and Mismeasure of Fairness
- Fairness in Machine Learning
- Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
- The Frontiers of Fairness in Machine Learning
- Flexibly Fair Representation Learning by Disentanglement
- Censoring Representations with an Adversary
- Equal Opportunity and Affirmative Action via Counterfactual Predictions