Optimal Transport for Fairness: Archival Data Repair using Small Research Data Sets

arXiv:2403.13864v1 Announce Type: new
Abstract: With the advent of the AI Act and other regulations, there is now an urgent need for algorithms that repair unfairness in training data. In this paper, we define fairness in terms of conditional independence between protected attributes ($S$) and features ($X$), given unprotected attributes ($U$). We address the important setting in which torrents of archival data need to be repaired, using only a small proportion of these data, which are $S|U$-labelled (the research data). …

abstract act ai act algorithms arxiv cs.cy cs.lg data data sets fairness features math.st paper regulations research small stat.th terms training training data transport type