Fairness Guarantees under Demographic Shift
Fairness Guarantees under Demographic Shift
复制标题
DOI：
10.4108/eai.12-1-2024.2347147
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Stephen Giguere;Blossom Metevier;B. C. Silva;Yuriy Brun;P. Thomas;S. Niekum
中科院分区：
文献类型：
--
作者：
Stephen Giguere;Blossom Metevier;B. C. Silva;Yuriy Brun;P. Thomas;S. Niekum
最近的研究发现，将机器学习用于社交应用可能会导致不公正，表现为种族主义、性别歧视，以及其他不公平和歧视性的结果。为了应对这一挑战，最近的机器学习算法已经设计成限制这种不公平行为发生的可能性。然而，这些方法通常假设用于培训的数据代表部署中将遇到的情况，这通常是不正确的。特别是，如果人口的某些子组在部署中或多或少变得可能(我们称之为人口结构变化)，以前的工作的公平性保证往往是无效的。在本文中，我们考虑了人口结构变化的影响，并提出了一类称为Shifty算法的算法，当来自部署环境的数据在训练期间不可用时，该算法在人口结构变化下提供高保真的行为保证。Shift是同类技术中的第一项，它展示了一种有效的算法设计策略，以克服人口结构变化带来的挑战。我们使用UCI成人人口普查数据集(Kohavi和Becker，1996)以及真实世界的大学入学考试和随后的学生成功数据集来评估Shifty。研究表明，与现有方法不同，所学习的模型避免了人口结构变化下的偏差。实验表明，该算法的高置信度公平性保证在实践中是有效的，并且该算法是在人口结构发生变化时训练公平模型的有效工具。
Recent studies found that using machine learning for social applications can lead to injustice in the form of racist, sexist, and otherwise unfair and discriminatory outcomes. To address this challenge, recent machine learning algorithms have been designed to limit the likelihood such unfair behavior occurs. However, these approaches typically assume the data used for training is representative of what will be encountered in deployment, which is often untrue. In particular, if certain subgroups of the population become more or less probable in deployment (a phenomenon we call demographic shift ), prior work’s fairness assurances are often invalid. In this paper, we consider the impact of demographic shift and present a class of algorithms, called Shifty algorithms, that provide high-con-fidence behavioral guarantees that hold under demographic shift when data from the deployment environment is unavailable during training. Shifty , the first technique of its kind, demonstrates an effective strategy for designing algorithms to overcome demographic shift’s challenges. We evaluate Shifty using the UCI Adult Census dataset (Kohavi and Becker, 1996), as well as a real-world dataset of university entrance exams and subsequent student success. We show that the learned models avoid bias under demographic shift, unlike existing methods. Our experiments demonstrate that our algorithm’s high-confidence fairness guarantees are valid in practice and that our algorithm is an effective tool for training models that are fair when demographic shift occurs.