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

TL;DR

This paper introduces an iterative orthogonal projection method to interpret black-box models, enabling quantification of input attribute importance to assess potential bias and discrimination.

Contribution

The paper proposes a novel iterative orthogonal projection technique for interpreting black-box models and quantifying input dependence to evaluate fairness.

Findings

Effective in quantifying attribute dependence

Helps identify potential bias in models

Provides a tool for fairness assessment

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.