多敏感特征下实现最优公平分类的新框架，突破单特征局限。
Bayes-Optimal Fair Classification with Multiple Sensitive Features
- 基于多敏感特征构建贝叶斯最优公平分类器，引入加权群体概率阈值规则。
- 在均值差与均值比等度量下，公平性可转化为特定群体选择率的线性变换。
- 适用于属性感知与盲设场景，支持如平等机会等复合公平准则。
现有贝叶斯最优公平分类的理论研究通常仅考虑单一（二元）敏感特征。现实中，个体常由多个敏感特征定义。本文在一般近似公平度量（包括均值差与均值比）下，刻画了多敏感特征情形下的贝叶斯最优公平分类器。我们证明，现有群体公平性概念（如人口均等、平等机会、预测均等、准确率均等）的近似度量，均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。进一步，我们揭示贝叶斯最优公平分类器转化为依赖于这些群体成员概率加权和的实例相关阈值规则。该框架适用于属性感知与属性盲设场景，并可处理如平等几率等复合公平性概念。在此基础上，我们提出两种基于内处理与后处理的贝叶斯最优公平分类实用算法。实验表明，所提方法在性能上优于现有方法。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。