Fair Kernel Regression via Fair Feature Embedding in Kernel Space
Fair Kernel Regression via Fair Feature Embedding in Kernel Space
复制标题
DOI：
10.1109/ictai.2019.00200
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Austin Okray;Hui Hu;Chao Lan
中科院分区：
文献类型：
--
作者：
Austin Okray;Hui Hu;Chao Lan
近年来，在减轻机器学习方法中不道德的人口统计偏差方面做出了重大努力。然而，针对内核方法所做的工作很少。本文提出了一种基于核空间公平特征嵌入的公平核回归方法(FKR-F^2E)。在前人研究公平学习的特征处理和核方法的特征选择的基础上，我们提出在核空间学习公平的特征嵌入，使特征分布的人口统计差异最小化。通过在三个公开的真实世界数据集上的实验表明，与现有的公平核回归方法和其他几种基线方法相比，所提出的FKR-F^2E具有更低的预测差异。
In recent years, there have been significant efforts on mitigating unethical demographic biases in machine learning methods. However, very little work is done for kernel methods. In this paper, we propose a novel fair kernel regression method via fair feature embedding (FKR-F^2E) in kernel space. Motivated by prior works feature processing for fair learning and feature selection for kernel methods, we propose to learn fair feature embeddings in kernel space, where the demographic discrepancy of feature distributions is minimized. Through experiments on three public real-world data sets, we show the proposed FKR-F^2E achieves significantly lower prediction disparity compared with the state-of-the-art fair kernel regression method and several other baseline methods.