Convex fairness constrained model using causal effect estimators
Convex fairness constrained model using causal effect estimators
复制标题
使用因果效应估计器的凸公平约束模型
DOI：
10.1145/3366424.3383556
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Akiko Takeda
中科院分区：
文献类型：
--
作者：
Hikaru Ogura;Akiko Takeda
近年来，人们对机器学习中的公平性进行了大量研究。在这里，平均差异（MD）或人口均等是最流行的公平措施之一。然而，MD不仅量化了歧视，还量化了解释性偏差，即由解释性特征证明的结果差异。在本文中，我们设计了新的模型，称为FairCEEs，消除歧视，同时保持解释性偏见。该模型是基于估计的因果关系的影响，利用倾向得分分析。我们证明了FairCEEs的平方损失理论上优于一个天真的MD约束模型。我们提供了一个有效的算法来解决FairCEE回归和二进制分类任务。在我们对这两项任务中的合成和真实数据进行的实验中，FairCEE的表现优于现有的模型，该模型考虑了特定情况下的解释性偏差。
Recent years have seen much research on fairness in machine learning. Here, mean difference (MD) or demographic parity is one of the most popular measures of fairness. However, MD quantifies not only discrimination but also explanatory bias which is the difference of outcomes justified by explanatory features. In this paper, we devise novel models, called FairCEEs, which remove discrimination while keeping explanatory bias. The models are based on estimators of causal effect utilizing propensity score analysis. We prove that FairCEEs with the squared loss theoretically outperform a naive MD constraint model. We provide an efficient algorithm for solving FairCEEs in regression and binary classification tasks. In our experiment on synthetic and real-world data in these two tasks, FairCEEs outperformed an existing model that considers explanatory bias in specific cases.