期刊
出版社
ASSOC COMPUTING MACHINERY
DOI: 10.1145/3404835.3462850
关键词
ranking algorithms; demographic inference; algorithmic fairness; ethical ai; noisy protected attributes; uncertainty
资金
- Sloan Fellowship award, NSF [IIS 1917668, IIS 1822831]
- Dow Chemical
This study investigates the impact of uncertainty and errors in demographic inference on the fairness of fair ranking algorithms. Through simulations and three case studies using real datasets, it demonstrates how demographic inferences drawn from real systems can result in unfair rankings. The results suggest that developers should refrain from using inferred demographic data as input to fair ranking algorithms unless the inferences are highly accurate.
Existing fair ranking systems, especially those designed to be demographically fair, assume that accurate demographic information about individuals is available to the ranking algorithm. In practice, however, this assumption may not hold - in real-world contexts like ranking job applicants or credit seekers, social and legal barriers may prevent algorithm operators from collecting peoples' demographic information. In these cases, algorithm operators may attempt to infer peoples' demographics and then supply these inferences as inputs to the ranking algorithm. In this study, we investigate how uncertainty and errors in demographic inference impact the fairness offered by fair ranking algorithms. Using simulations and three case studies with real datasets, we show how demographic inferences drawn from real systems can lead to unfair rankings. Our results suggest that developers should not use inferred demographic data as input to fair ranking algorithms, unless the inferences are extremely accurate.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据