Bayes-Optimal Fair Classification with Linear Disparity Constraints via Pre-, In-, and Post-processing
Department of Statistics and Data Science, Xiamen University(厦门大学统计与数据科学系);Department of Statistics and Data Science, University of Pennsylvania(宾夕法尼亚大学统计与数据科学系);Dept of Statistics and Data Science, University of California, Los Angeles(加州大学洛杉矶分校统计与数据科学系)
Xianli Zeng, Kevin Jiang, Guang Cheng, Edgar Dobriban
备注:
This paper replaces the preprint "Bayes-optimal classifiers under group fairness" by Xianli Zeng, Edgar Dobriban, and Guang Cheng (arXiv:2202.09724)
Machine learning algorithms may have disparate impacts on protected groups. To address this, we develop methods for Bayes-optimal fair classification, aiming to minimize classification error subject to given group fairness constraints. We introduce the notion of \emph{linear disparity measures}, which are linear functions of a probabilistic classifier; and \emph{bilinear disparity measures}, which are also linear in the group-wise regression functions. We show that several popular disparity measures -- the deviations from demographic parity, equality of opportunity, and predictive equality -- are bilinear.
We find the form of Bayes-optimal fair classifiers under a single linear disparity measure, by uncovering a connection with the Neyman-Pearson lemma. For bilinear disparity measures, we are able to find the explicit form of Bayes-optimal fair classifiers as group-wise thresholding rules with explicitly characterized thresholds. We develop similar algorithms for when protected attribute cannot be used at the prediction phase. Moreover, we obtain analogous theoretical characterizations of optimal classifiers for a multi-class protected attribute and for equalized odds.
Leveraging our theoretical results, we design methods that learn fair Bayes-optimal classifiers under bilinear disparity constraints. Our methods cover three popular approaches to fairness-aware classification, via pre-processing (Fair Up- and Down-Sampling), in-processing (Fair cost-sensitive Classification) and post-processing (a Fair Plug-In Rule). Our methods control disparity directly while achieving near-optimal fairness-accuracy tradeoffs. We show empirically that our methods have state-of-the-art performance compared to existing algorithms. In particular, our pre-processing method can a reach higher accuracy than prior pre-processing methods at low disparity levels.
Group Retention when Using Machine Learning in Sequential Decision Making: the Interplay between User Dynamics and Fairness
University of Michigan(密歇根大学);Bilkent University(比尔肯特大学)
在来自多个人口统计组的数据上训练的机器学习(ML)模型可以继承表示差异(Hashimoto et al.,数据中可能存在的问题:该模型可能对对培训过程贡献较少的群体不太有利;这反过来又会随着时间的推移降低这些群体的人口保留率,并从长远来看加剧代表性差异。在这项研究中,我们试图了解ML决策和底层群体表示之间的相互作用,它们如何在顺序框架中演变,以及公平标准的使用如何在这一过程中发挥作用。我们表明,代表性的差距可以很容易地恶化,随着时间的推移,在一个自然的用户动态(到达和离开)模型的基础上,一个常用的客观和公平的标准作出决定时,从长远来看,导致一些群体完全从样本池减少。它强调了这样一个事实,即必须定义公平性标准,同时考虑到决策对用户动态的影响。为此,我们解释了如何选择一个适当的公平性标准的基础上,一般的用户动态模型。
Bayes-Optimal Fair Classification with Multiple Sensitive Features
Department of Information Systems, Arizona State University(信息系统系,亚利桑那州立大学);Department of Information Systems and Intelligent Business, School of Management, Xi’an Jiaotong University(信息系统与智能商业系,管理学院,西安交通大学)
PAC-Bayesian Generalization Guarantees for Fairness on Stochastic and Deterministic Classifiers
CNRS(法国国家科学研究中心);Institut d Optique Graduate School(高等光学学院);Inria(法国国家信息与自动化研究所);Institut Universitaire de France(法国大学研究院);Universite Claude Bernard Lyon 1(里昂第一大学);Univ Rennes(雷恩大学)
Penalized Fair Regression for Multiple Groups in Chronic Kidney Disease
Department of Health Policy, Stanford University(健康政策系,斯坦福大学);Stanford University(斯坦福大学);Department of Epidemiology and Population Health, Stanford University(流行病学与人口健康系,斯坦福大学)