Iterative Orthogonal Feature Projection for Diagnosing Bias in Black-Box Models
Iterative Orthogonal Feature Projection for Diagnosing Bias in Black-Box Models
复制标题
用于诊断黑盒模型偏差的迭代正交特征投影
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Lalana Kagal
中科院分区：
文献类型：
--
作者：
Julius Adebayo;Lalana Kagal
预测模型越来越多地用于确定获得信贷，保险和就业等服务的目的。尽管生产力和效率有可能提高，但一些潜在的问题尚未得到解决，特别是无意歧视的可能性。我们提出了一个迭代过程，基于正交投影的输入属性，使黑箱预测模型的可解释性。通过我们的迭代过程，可以量化黑箱模型对输入属性的相对依赖性，然后可以使用输入对预测模型的相对重要性来评估这种模型的公平性（或歧视程度）。
Predictive models are increasingly deployed for the purpose of determining access to services such as credit, insurance, and employment. Despite potential gains in productivity and efficiency, several potential problems have yet to be addressed, particularly the potential for unintentional discrimination. We present an iterative procedure, based on orthogonal projection of input attributes, for enabling interpretability of black-box predictive models. Through our iterative procedure, one can quantify the relative dependence of a black-box model on its input attributes.The relative significance of the inputs to a predictive model can then be used to assess the fairness (or discriminatory extent) of such a model.