arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.05049stat.MLcs.LG
收藏

训练时公平,部署时不公平:可观察的公平性测量在执行预测设置中不稳定

Fair When Trained, Unfair When Deployed: Observable Fairness Measures are Unstable in Performative Prediction Settings

J.P. Morgan AI Research(摩根大通人工智能研究)

Alan Mishler, Niccolò Dalmasso

摘要:

许多流行的算法公平性度量依赖于预测、结果和种族或性别等敏感特征的联合分布。这些措施是敏感的分布转移:一个预测器,这是训练,以满足这些公平的定义之一,可能会变得不公平,如果分布变化。然而,在表演性预测环境中,预测因子的目的恰恰是诱导分布转移。例如,在刑事司法、医疗保健和消费金融的许多应用中,构建预测器的目的是降低不良结果的发生率,如累犯、住院或贷款违约。我们形式化的影响,这种预测作为一种类型的概念转变,一个特定的各种分布转变,并显示理论上和通过模拟的例子,这是如何导致预测,这是公平的,当他们被训练成为不公平的,当他们被部署。我们进一步展示了这些问题中有多少可以通过使用依赖于反事实而不是可观察结果的公平定义来避免。

英文摘要:

Many popular algorithmic fairness measures depend on the joint distribution of predictions, outcomes, and a sensitive feature like race or gender. These measures are sensitive to distribution shift: a predictor which is trained to satisfy one of these fairness definitions may become unfair if the distribution changes. In performative prediction settings, however, predictors are precisely intended to induce distribution shift. For example, in many applications in criminal justice, healthcare, and consumer finance, the purpose of building a predictor is to reduce the rate of adverse outcomes such as recidivism, hospitalization, or default on a loan. We formalize the effect of such predictors as a type of concept shift-a particular variety of distribution shift-and show both theoretically and via simulated examples how this causes predictors which are fair when they are trained to become unfair when they are deployed. We further show how many of these issues can be avoided by using fairness definitions that depend on counterfactual rather than observable outcomes.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑