Existing theoretical work on Bayes-optimal fair classifiers usually considers a single (binary) sensitive feature. In practice, individuals are often defined by multiple sensitive features. In this paper, we characterize the Bayes-optimal fair classifier for multiple sensitive features under general approximate fairness measures, including mean difference and mean ratio. We show that these approximate measures for existing group fairness notions, including Demographic Parity, Equal Opportunity, Predictive Equality, and Accuracy Parity, are linear transformations of selection rates for specific groups defined by both labels and sensitive features. We then characterize that Bayes-optimal fair classifiers for multiple sensitive features become instance-dependent thresholding rules that rely on a weighted sum of these group membership probabilities. Our framework applies to both attribute-aware and attribute-blind settings and can accommodate composite fairness notions like Equalized Odds. Building on this, we propose two practical algorithms for Bayes-optimal fair classification via in-processing and post-processing. We show empirically that our methods compare favorably to existing methods.
翻译:现有关于贝叶斯最优公平分类器的理论研究通常仅考虑单一(二元)敏感特征。在实践中,个体往往由多重敏感特征定义。本文中,我们刻画了在多重敏感特征下、基于一般近似公平度量(包括均值差异与均值比率)的贝叶斯最优公平分类器。我们证明,对于现有群体公平性概念——包括人口统计均等、机会均等、预测均等与准确率均等——这些近似度量可表示为由标签与敏感特征共同定义的特定群体选择率的线性变换。进而,我们刻画了多重敏感特征下的贝叶斯最优公平分类器,其表现为依赖于这些群体隶属概率加权和的实例依赖型阈值规则。本框架适用于属性感知与属性未知两种设置,并可容纳如均衡几率等复合公平性概念。基于此,我们提出两种通过处理中与处理后实现贝叶斯最优公平分类的实用算法。实验表明,我们的方法优于现有方法。