approved
Algorithmic decision making and the cost of fairness

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.

Tags
Data and Resources
To access the resources you must log in
  • BibTeXBibTeX

    The resource: 'BibTeX' is not accessible as guest user. You must login to access it!
  • htmlHTML

    The resource: 'html' is not accessible as guest user. You must login to access it!
Additional Info
Field Value
Author Pierson, Emma
Author Feller, Avi
Author Goel, Sharad
Author Huq, Aziz
Author Corbett Davies, Sam, scorbe@stanford.edu
DOI https://doi.org/10.1145/3097983.3098095
Group Ethics and Legality
Publisher ACM
Source KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAugust, 2017, Pages 797–806
Thematic Cluster Other
system:type ConferencePaper
Management Info
Field Value
Author Pozzi Giorgia
Maintainer Pozzi Giorgia
Version 1
Last Updated 19 July 2022, 16:32 (CEST)
Created 9 February 2021, 13:41 (CET)