Algorithmic Decision Making and the Cost of Fairness

data_society's bookmarks 2019-08-20

Type Conference Paper Author Sam Corbett-Davies Author Emma Pierson Author Avi Feller Author Sharad Goel Author Aziz Huq URL http://doi.acm.org/10.1145/3097983.3098095 Series KDD '17 Place New York, NY, USA Publisher ACM Pages 797–806 ISBN 978-1-4503-4887-4 Date 2017 Extra event-place: Halifax, NS, Canada DOI 10.1145/3097983.3098095 Accessed 2019-06-21 16:39:48 Library Catalog ACM Digital Library Abstract Algorithms are now regularly used to decide whether defendants awaiting trial are too dangerous to be released back into the community. In some cases, black defendants are substantially more likely than white defendants to be incorrectly classified as high risk. To mitigate such disparities, several techniques have recently been proposed to achieve algorithmic fairness. Here we reformulate algorithmic fairness as constrained optimization: the objective is to maximize public safety while satisfying formal fairness constraints designed to reduce racial disparities. We show that for several past definitions of fairness, the optimal algorithms that result require detaining defendants above race-specific risk thresholds. We further show that the optimal unconstrained algorithm requires applying a single, uniform threshold to all defendants. The unconstrained algorithm thus maximizes public safety while also satisfying one important understanding of equality: that all individuals are held to the same standard, irrespective of race. Because the optimal constrained and unconstrained algorithms generally differ, there is tension between improving public safety and satisfying prevailing notions of algorithmic fairness. By examining data from Broward County, Florida, we show that this trade-off can be large in practice. We focus on algorithms for pretrial release decisions, but the principles we discuss apply to other domains, and also to human decision makers carrying out structured decision rules. Proceedings Title Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining