Computer Science
Supervised Algorithmic Fairness in Distribution Shifts: A Survey
M. Shao, D. Li, et al.
How can fairness survive when data distributions shift? This survey, conducted by Minglai Shao, Dong Li, Chen Zhao, Xintao Wu, Yujie Lin, and Qin Tian, summarizes types of distribution shifts (covariate, label, concept, demographic, dependence), reviews six approaches (e.g., disentanglement, causal methods, reweighting, robust optimization, regularization), lists datasets and metrics, and outlines challenges and future directions—an essential guide for fairness-aware models under real-world change.
Related Publications
Explore these studies to deepen your understanding of the subject.
