Optimal Group Fair Classifiers from Linear Post-Processing

arXiv:2405.04025v1 Announce Type: new
Abstract: We propose a post-processing algorithm for fair classification that mitigates model bias under a unified family of group fairness criteria covering statistical parity, equal opportunity, and equalized odds, applicable to multi-class problems and both attribute-aware and attribute-blind settings. It achieves fairness by re-calibrating the output score of the given base model with a "fairness cost" -- a linear combination of the (predicted) group memberships. Our algorithm is based on a representation result showing that the …

abstract algorithm arxiv bias blind class classification classifiers cs.cy cs.lg equal fair fairness family linear model bias post-processing processing statistical type