To Pool or Not To Pool: Analyzing the Regularizing Effects of Group-Fair Training on Shared Models

arXiv:2402.18803v1 Announce Type: new
Abstract: In fair machine learning, one source of performance disparities between groups is over-fitting to groups with relatively few training samples. We derive group-specific bounds on the generalization error of welfare-centric fair machine learning that benefit from the larger sample size of the majority group. We do this by considering group-specific Rademacher averages over a restricted hypothesis class, which contains the family of models likely to perform well with respect to a fair learning objective (e.g., …

abstract arxiv benefit cs.cy cs.lg effects error fair machine machine learning performance pool sample samples training type welfare