Iterative Orthogonal Feature Projection for Diagnosing Bias in Black-Box Models

Abstract

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.

References in corpus (1)

• Certifying and removing disparate impact

Cited by in corpus (13)

• Fairness Testing: Testing Software for Discrimination
• Fairness in Machine Learning
• Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation
• Do the Machine Learning Models on a Crowd Sourced Platform Exhibit Bias? An Empirical Study on Model Fairness
• Empirical Risk Minimization under Fairness Constraints
• Transforming Feature Space to Interpret Machine Learning Models
• Automated Test Generation to Detect Individual Discrimination in AI Models
• The Authority of "Fair" in Machine Learning
• Kernel Dependence Regularizers and Gaussian Processes with Applications to Algorithmic Fairness
• Fairkit, Fairkit, on the Wall, Who's the Fairest of Them All? Supporting Data Scientists in Training Fair Models
• Algorithm Fairness in AI for Medicine and Healthcare
• Metamorphic Testing of a Deep Learning based Forecaster
• On the Direction of Discrimination: An Information-Theoretic Analysis of Disparate Impact in Machine Learning