中文

回归背景下一种满足人口均等且均衡组间风险的预测示例

机器学习 2020-11-17 v1 人工智能 机器学习

摘要

(X,S,Y)Rp×{1,2}×R(X, S, Y) \in \mathbb{R}^p \times \{1, 2\} \times \mathbb{R} 为遵循某联合分布 P\mathbb{P} 的三元组,其中 XX 为特征向量,SS 为敏感属性,YY 为目标变量。不产生差别对待的贝叶斯最优预测 ff^* 定义为 f(x)=E[YX=x]f^*(x) = \mathbb{E}[Y | X = x]。我们给出一个非平凡的预测 xf(x)x \to f(x) 示例,其满足两种常见的组公平概念:人口均等(Demographic Parity)\begin{align} (f(X) | S = 1) &\stackrel{d}{=} (f(X) | S = 2) \end{align} 与均衡组间风险(Equal Group-Wise Risks)\begin{align} \mathbb{E}[(f^*(X) - f(X))^2 | S = 1] = \mathbb{E}[(f^*(X) - f(X))^2 | S = 2]. \end{align} 据我们所知,这是首个满足上述条件的非常数预测器的显式构造。我们讨论了该结果对更好理解算法公平数学概念的若干启示。

关键词

引用

@article{arxiv.2011.07158,
  title  = {An example of prediction which complies with Demographic Parity and equalizes group-wise risks in the context of regression},
  author = {Evgenii Chzhen and Nicolas Schreuder},
  journal= {arXiv preprint arXiv:2011.07158},
  year   = {2020}
}

备注

Presented at the NeurIPS 2020 Workshop on Algorithmic Fairness through the Lens of Causality and Interpretability