When Fair Classification Meets Noisy Protected Attributes
This addresses the practical challenge of implementing fair classification in real-world scenarios where demographic data is unreliable or legally restricted, though it is incremental as it compares existing algorithm types.
The study tackled the problem of ensuring algorithmic fairness when protected attributes are noisy or unavailable by comparing attribute-reliant, noise-tolerant, and attribute-blind fair classification algorithms. It found that attribute-blind and noise-tolerant methods can achieve similar performance and fairness levels as attribute-reliant ones, even with noisy attributes, based on evaluations on four real-world datasets and synthetic perturbations.
The operationalization of algorithmic fairness comes with several practical challenges, not the least of which is the availability or reliability of protected attributes in datasets. In real-world contexts, practical and legal impediments may prevent the collection and use of demographic data, making it difficult to ensure algorithmic fairness. While initial fairness algorithms did not consider these limitations, recent proposals aim to achieve algorithmic fairness in classification by incorporating noisiness in protected attributes or not using protected attributes at all. To the best of our knowledge, this is the first head-to-head study of fair classification algorithms to compare attribute-reliant, noise-tolerant and attribute-blind algorithms along the dual axes of predictivity and fairness. We evaluated these algorithms via case studies on four real-world datasets and synthetic perturbations. Our study reveals that attribute-blind and noise-tolerant fair classifiers can potentially achieve similar level of performance as attribute-reliant algorithms, even when protected attributes are noisy. However, implementing them in practice requires careful nuance. Our study provides insights into the practical implications of using fair classification algorithms in scenarios where protected attributes are noisy or partially available.