Learning Fair and Interpretable Representations via Linear Orthogonalization
arXiv:1910.12854
Abstract
To reduce human error and prejudice, many high-stakes decisions have been turned over to machine algorithms. However, recent research suggests that this does not remove discrimination, and can perpetuate harmful stereotypes. While algorithms have been developed to improve fairness, they typically face at least one of three shortcomings: they are not interpretable, their prediction quality deteriorates quickly compared to unbiased equivalents, and they are not easily transferable across models. To address these shortcomings, we propose a geometric method that removes correlations between data and any number of protected variables. Further, we can control the strength of debiasing through an adjustable parameter to address the trade-off between prediction quality and fairness. The resulting features are interpretable and can be used with many popular models, such as linear regression, random forest, and multilayer perceptrons. The resulting predictions are found to be more accurate and fair compared to several state-of-the-art fair AI algorithms across a variety of benchmark datasets. Our work shows that debiasing data is a simple and effective solution toward improving fairness.
9 pages, 5 figures
References in corpus (11)
- Fairness Constraints: Mechanisms for Fair Classification
- The Frontiers of Fairness in Machine Learning
- A Convex Framework for Fair Regression
- The Variational Fair Autoencoder
- Why Is My Classifier Discriminatory?
- From Parity to Preference-based Notions of Fairness in Classification
- Invariant Representations without Adversarial Training
- Unsupervised Adversarial Invariance
- Learning unbiased features
- Unified Adversarial Invariance
- On the Fairness of Disentangled Representations