Fair Classification under Covariate Shift and Missing Protected Attribute -- an Investigation using Related Features
arXiv:2204.07987v1h-index: 2
Originality Synthesis-oriented
AI Analysis
This addresses fairness in machine learning for scenarios with data distribution shifts and incomplete sensitive information, but it appears incremental as it combines existing techniques.
The study tackled fair classification under covariate shift and missing protected attributes by using importance weights and related features, achieving unspecified results without concrete numbers.
This study investigated the problem of fair classification under Covariate Shift and missing protected attribute using a simple approach based on the use of importance-weights to handle covariate-shift and, Related Features arXiv:2104.14537 to handle missing protected attribute.