arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2002.02901stat.MLcs.LGmath.STstat.TH
收藏

不经意的数据与核的公平性

Oblivious Data for Fairness with Kernels

Lancaster University(兰卡斯特大学)

Steffen Grünewälder, Azadeh Khaleghi

摘要:

我们调查的问题,算法的公平性的情况下,敏感和非敏感的功能是可用的,一个目的是生成新的,“遗忘”,功能,密切近似的非敏感功能,只有最低限度地依赖于敏感的。我们在核方法的上下文中研究这个问题。我们分析了一个宽松的版本的最大均值离散准则,不保证完全独立,但使优化问题易于处理。我们推导出一个封闭形式的解决方案,这个放松的优化问题,并补充与新生成的功能和敏感的研究之间的依赖关系的结果。我们生成这种不经意特征的关键因素是希尔伯特空间值条件期望,需要从数据中估计。我们提出了一个插件的方法,并演示了如何可以控制的估计误差。虽然我们的技术有助于减少偏差,但我们想指出的是,任何数据集的后处理都不可能替代精心设计的实验。

英文摘要:

We investigate the problem of algorithmic fairness in the case where sensitive and non-sensitive features are available and one aims to generate new, `oblivious', features that closely approximate the non-sensitive features, and are only minimally dependent on the sensitive ones. We study this question in the context of kernel methods. We analyze a relaxed version of the Maximum Mean Discrepancy criterion which does not guarantee full independence but makes the optimization problem tractable. We derive a closed-form solution for this relaxed optimization problem and complement the result with a study of the dependencies between the newly generated features and the sensitive ones. Our key ingredient for generating such oblivious features is a Hilbert-space-valued conditional expectation, which needs to be estimated from data. We propose a plug-in approach and demonstrate how the estimation errors can be controlled. While our techniques help reduce the bias, we would like to point out that no post-processing of any dataset could possibly serve as an alternative to well-designed experiments.

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。

↑