Convergent Algorithms for (Relaxed) Minimax Fairness
Convergent Algorithms for (Relaxed) Minimax Fairness
复制标题
（宽松）极小极大公平性的收敛算法
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Aaron Roth
中科院分区：
文献类型：
--
作者：
Emily Diana;Wesley Gill;Michael Kearns;K. Kenthapadi;Aaron Roth
我们考虑一个最近引入的框架，在该框架中，公平性是通过跨组的最坏情况下的结果来衡量的，而不是通过组结果之间更标准的$\textit{difference}$。在此框架中，我们为$\textit{minimax group fairness}$提供可证明收敛的$\textit{oracle-efficient}$学习算法（或等效地，减少非公平学习）。这里的目标是最小化所有组的最大损失，而不是均衡组损失。我们的算法适用于回归和分类设置，并支持整体错误率和假阳性或假阴性率作为兴趣的公平性度量。他们还支持放宽公平性约束，从而允许研究的整体准确性和极小极大公平之间的权衡。我们在各种公平敏感的数据集上比较了我们的算法的实验行为和性能，并显示了极大极小公平性严格且强烈优于平等结果概念的情况，在这个意义上，平等结果只能通过人为地夸大对某些群体造成的伤害来获得，而不是他们在极大极小解决方案下遭受的伤害。
We consider a recently introduced framework in which fairness is measured by worst-case outcomes across groups, rather than by the more standard $\textit{difference}$ between group outcomes. In this framework we provide provably convergent $\textit{oracle-efficient}$ learning algorithms (or equivalently, reductions to non-fair learning) for $\textit{minimax group fairness}$. Here the goal is that of minimizing the maximum loss across all groups, rather than equalizing group losses. Our algorithms apply to both regression and classification settings and support both overall error and false positive or false negative rates as the fairness measure of interest. They also support relaxations of the fairness constraints, thus permitting study of the tradeoff between overall accuracy and minimax fairness. We compare the experimental behavior and performance of our algorithms across a variety of fairness-sensitive data sets and show cases in which minimax fairness is strictly and strongly preferable to equal outcome notions, in the sense that equal outcomes can only be obtained by artificially inflating the harm inflicted on some groups compared to what they suffer under the minimax solution.
DOI：
--
发表时间：
2020-07
期刊：
Proceedings of machine learning research
影响因子：
--
作者：
Natalia Martínez;Martín Bertrán;G. Sapiro
通讯作者：
Natalia Martínez;Martín Bertrán;G. Sapiro
DOI：
--
发表时间：
2018-10
期刊：
--
影响因子：
--
作者：
S. Samadi;U. Tantipongpipat;Jamie Morgenstern;Mohit Singh;S. Vempala
通讯作者：
S. Samadi;U. Tantipongpipat;Jamie Morgenstern;Mohit Singh;S. Vempala