arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.14263stat.MLcs.CYcs.LG
收藏

通过成人数据集的统计奇偶性棱镜对机器学习中的偏见进行调查

A survey of bias in Machine Learning through the prism of Statistical Parity for the Adult Data Set

Institut de Mathématiques de Toulouse(图卢兹数学研究所);INSA(国立应用科学学院);Université Toulouse 3(图卢兹第三大学);CNRS(法国国家科学研究中心);Instituto de Matemáticas de la Universidad de Valladolid(巴利亚多利德大学数学研究所);Universidad de Valladolid(巴利亚多利德大学)

Philippe Besse, Eustasio del Barrio, Paula Gordaliza, Jean-Michel Loubes, Laurent Risser

摘要:

基于机器学习模型的应用程序现已成为日常生活和专业领域不可或缺的一部分。一个关键的问题最近在人群中出现:算法决策是否传达了对特定人群或少数民族的任何类型的歧视?在本文中,我们展示了理解如何将偏差引入自动决策的重要性。我们首先提出了一个数学框架的公平学习问题,特别是在二进制分类设置。然后,我们建议量化存在的偏见,通过使用标准的差异影响指数的真实和众所周知的成人收入数据集。最后,我们检查了不同方法的性能,旨在减少二进制分类结果的偏差。重要的是,我们表明,一些直观的方法是无效的。这揭示了一个事实,即试图建立公平的机器学习模型可能是一项特别具有挑战性的任务,特别是当训练观察结果包含偏差时。

英文摘要:

Applications based on Machine Learning models have now become an indispensable part of the everyday life and the professional world. A critical question then recently arised among the population: Do algorithmic decisions convey any type of discrimination against specific groups of population or minorities? In this paper, we show the importance of understanding how a bias can be introduced into automatic decisions. We first present a mathematical framework for the fair learning problem, specifically in the binary classification setting. We then propose to quantify the presence of bias by using the standard Disparate Impact index on the real and well-known Adult income data set. Finally, we check the performance of different approaches aiming to reduce the bias in binary classification outcomes. Importantly, we show that some intuitive methods are ineffective. This sheds light on the fact trying to make fair machine learning models may be a particularly challenging task, in particular when the training observations contain a bias.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑