Fair Active Learning
This work addresses fairness in machine learning for societal applications, but it is incremental as it applies existing fairness measures to active learning.
The paper tackles the problem of building accurate and fair classifiers under labeling budget constraints by designing fair active learning algorithms that balance model accuracy and demographic parity, demonstrating effectiveness through extensive experiments on benchmark datasets.
Machine learning (ML) is increasingly being used in high-stakes applications impacting society. Therefore, it is of critical importance that ML models do not propagate discrimination. Collecting accurate labeled data in societal applications is challenging and costly. Active learning is a promising approach to build an accurate classifier by interactively querying an oracle within a labeling budget. We design algorithms for fair active learning that carefully selects data points to be labeled so as to balance model accuracy and fairness. Specifically, we focus on demographic parity - a widely used measure of fairness. Extensive experiments over benchmark datasets demonstrate the effectiveness of our proposed approach.