We extend the notion of minimax fairness in supervised learning problems to its natural conclusion: lexicographic minimax fairness (or lexifairness for short). Informally, given a collection of demographic groups of interest, minimax fairness asks that the error of the group with the highest error be minimized. Lexifairness goes further and asks that amongst all minimax fair solutions, the error of the group with the second highest error should be minimized, and amongst all of those solutions, the error of the group with the third highest error should be minimized, and so on. Despite its naturalness, correctly defining lexifairness is considerably more subtle than minimax fairness, because of inherent sensitivity to approximation error. We give a notion of approximate lexifairness that avoids this issue, and then derive oracle-efficient algorithms for finding approximately lexifair solutions in a very general setting. When the underlying empirical risk minimization problem absent fairness constraints is convex (as it is, for example, with linear and logistic regression), our algorithms are provably efficient even in the worst case. Finally, we show generalization bounds -- approximate lexifairness on the training sample implies approximate lexifairness on the true distribution with high probability. Our ability to prove generalization bounds depends on our choosing definitions that avoid the instability of naive definitions.
翻译:我们把受监督学习问题的最低公平概念扩大到其自然结论:字典小法公平(或短期法公平 ) 。 非正式地说,鉴于人口利益集团的集合,小法公平要求将错误降到最高差错最小。 利法公平更进一步,并要求在所有小型公平解决办法中,应尽量减少第二大差错群体在受监督学习问题上的错误,在所有这些解决办法中,应尽量减少第三大差错,等等。尽管该组的错误是自然的,正确界定法公平比小法公平要微妙得多,这是因为对近似差的内在敏感性。 我们给出了一个大致的法公平概念,避免了这一问题,然后得出了在非常笼统的环境下找到大致法公平解决办法的极效算法。 当基本的经验风险最小化问题不存在公平性制约时,我们的算法是相当有效的(例如,线性和后勤性倒退),即使最坏的情况,我们的算法也是非常有效的。 最后,我们展示了普遍法公平性的约束性,即我们所选择的高度法正统性定义的准确性,意味着我们所选择的精准性定义的精确性取决于一般法定义。