Recidivism prediction instruments (RPI's) provide decision makers with an
assessment of the likelihood that a criminal defendant will reoffend at a
future point in time. While such instruments are gaining increasing popularity
across the country, their use is attracting tremendous controversy. Much of the
controversy concerns potential discriminatory bias in the risk assessments that
are produced. This paper discusses several fairness criteria that have recently
been applied to assess the fairness of recidivism prediction instruments. We
demonstrate that the criteria cannot all be simultaneously satisfied when
recidivism prevalence differs across groups. We then show how disparate impact
can arise when a recidivism prediction instrument fails to satisfy the
criterion of error rate balance.

本文讨论了最近被应用于评估惯犯再犯预测工具公平性的几个公正标准，证明了当不同群体的惯犯再犯率不同时，这些标准无法同时满足。然后说明了当再犯预测工具未能满足错误率平衡标准时如何产生不同影响。

具有不同影响的公平预测：再犯罪预测工具中偏见研究

Fair prediction with disparate impact: A study of bias in recidivism  prediction instruments

The discovery of discriminatory bias in human or automated decision making is
a task of increasing importance and difficulty, exacerbated by the pervasive
use of machine learning and data mining. Currently, discrimination discovery
largely relies upon correlation analysis of decisions records, disregarding the
impact of confounding biases. We present a method for causal discrimination
discovery based on propensity score analysis, a statistical tool for filtering
out the effect of confounding variables. We introduce causal measures of
discrimination which quantify the effect of group membership on the decisions,
and highlight causal discrimination/favoritism patterns by learning regression
trees over the novel measures. We validate our approach on two real world
datasets. Our proposed framework for causal discrimination has the potential to
enhance the transparency of machine learning with tools for detecting
discriminatory bias both in the training data and in the learning algorithms.

本文介绍一种基于倾向得分分析的原因型歧视方法，并利用回归树学习新的歧视 / 偏袒模式，验证在两个现实世界的数据集上，它可以加强机器学习透明度并探测训练数据和学习算法中的歧视性偏差。