Measuring Non-Expert Comprehension of Machine Learning Fairness Metrics

TL;DR

This paper investigates whether laypeople understand key machine learning fairness definitions by developing a comprehension metric, evaluating it through an online survey, and analyzing factors influencing understanding.

Contribution

It introduces a novel metric to assess public understanding of ML fairness definitions and provides empirical insights into comprehension levels and influencing factors.

Findings

Lay understanding of fairness definitions varies significantly.

Demographic factors influence comprehension levels.

Sentiment towards fairness concepts correlates with understanding.

Abstract

Bias in machine learning has manifested injustice in several areas, such as medicine, hiring, and criminal justice. In response, computer scientists have developed myriad definitions of fairness to correct this bias in fielded algorithms. While some definitions are based on established legal and ethical norms, others are largely mathematical. It is unclear whether the general public agrees with these fairness definitions, and perhaps more importantly, whether they understand these definitions. We take initial steps toward bridging this gap between ML researchers and the public, by addressing the question: does a lay audience understand a basic definition of ML fairness? We develop a metric to measure comprehension of three such definitions--demographic parity, equal opportunity, and equalized odds. We evaluate this metric using an online survey, and investigate the relationship between comprehension and sentiment, demographics, and the definition itself.