A Fair Classifier Using Kernel Density Estimation
A Fair Classifier Using Kernel Density Estimation
复制标题
使用核密度估计的公平分类器
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Changho Suh
中科院分区：
文献类型：
--
作者：
Jaewoong Cho;Gyeongjo Hwang;Changho Suh
随着机器学习在招聘和刑事司法等一系列敏感应用中变得越来越普遍，机器学习分类器设计的一个关键方面是确保公平性：保证预测与性别和种族等敏感属性无关。这项工作开发了一种核密度估计（KDE）方法，忠实地尊重公平性约束，同时产生一个易于处理的优化问题，该问题具有高精度-公平性权衡。这种方法的一个关键特征是基于 KDE 量化的公平性度量可以表示为一个可微函数 w.r.t.模型参数，从而能够使用突出的梯度下降来轻松解决感兴趣的优化问题。这项工作的重点是分类任务和两个著名的群体公平性衡量标准：人口平等和均等赔率。我们的经验表明，我们的算法在准确性-公平性权衡以及合成数据集和基准真实数据集的训练稳定性方面比先前的公平分类器实现了更好或可比的性能。
As machine learning becomes prevalent in a widening array of sensitive applications such as job hiring and criminal justice, one critical aspect in the design of machine learning classifiers is to ensure fairness: Guaranteeing the irrelevancy of a prediction to sensitive attributes such as gender and race. This work develops a kernel density estimation (KDE) methodology to faithfully respect the fairness constraint while yielding a tractable optimization problem that comes with high accuracy-fairness tradeoff. One key feature of this approach is that the fairness measure quantified based on KDE can be expressed as a differentiable function w.r.t. model parameters, thereby enabling the use of prominent gradient descent to readily solve an interested optimization problem. This work focuses on classifi-cation tasks and two well-known measures of group fairness: demographic parity and equalized odds. We empirically show that our algorithm achieves greater or comparable performances against prior fair classifers in accuracy-fairness tradeoff as well as in training stability on both synthetic and benchmark real datasets.
DOI：
10.24963/ijcai.2019/199
发表时间：
2019-08
期刊：
--
影响因子：
--
作者：
Yongkai Wu;Lu Zhang;Xintao Wu
通讯作者：
Yongkai Wu;Lu Zhang;Xintao Wu