License: Creative Commons Attribution 3.0 Unported license (CC-BY 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.FORC.2020.6
URN: urn:nbn:de:0030-drops-120225
URL: https://drops.dagstuhl.de/opus/volltexte/2020/12022/
Go to the corresponding LIPIcs Volume Portal


Rambachan, Ashesh ; Roth, Jonathan

Bias In, Bias Out? Evaluating the Folk Wisdom

pdf-format:
LIPIcs-FORC-2020-6.pdf (0.5 MB)


Abstract

We evaluate the folk wisdom that algorithmic decision rules trained on data produced by biased human decision-makers necessarily reflect this bias. We consider a setting where training labels are only generated if a biased decision-maker takes a particular action, and so "biased" training data arise due to discriminatory selection into the training data. In our baseline model, the more biased the decision-maker is against a group, the more the algorithmic decision rule favors that group. We refer to this phenomenon as bias reversal. We then clarify the conditions that give rise to bias reversal. Whether a prediction algorithm reverses or inherits bias depends critically on how the decision-maker affects the training data as well as the label used in training. We illustrate our main theoretical results in a simulation study applied to the New York City Stop, Question and Frisk dataset.

BibTeX - Entry

@InProceedings{rambachan_et_al:LIPIcs:2020:12022,
  author =	{Ashesh Rambachan and Jonathan Roth},
  title =	{{Bias In, Bias Out? Evaluating the Folk Wisdom}},
  booktitle =	{1st Symposium on Foundations of Responsible Computing (FORC 2020)},
  pages =	{6:1--6:15},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-142-9},
  ISSN =	{1868-8969},
  year =	{2020},
  volume =	{156},
  editor =	{Aaron Roth},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2020/12022},
  URN =		{urn:nbn:de:0030-drops-120225},
  doi =		{10.4230/LIPIcs.FORC.2020.6},
  annote =	{Keywords: fairness, selective labels, discrimination, training data}
}

Keywords: fairness, selective labels, discrimination, training data
Collection: 1st Symposium on Foundations of Responsible Computing (FORC 2020)
Issue Date: 2020
Date of publication: 18.05.2020


DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI