# Fairness in representation: quantifying stereotyping as a representational harm

## TL;DR

This paper formalizes two types of stereotyping as representational harms in machine learning, demonstrating their impact on allocation and proposing mitigation strategies validated on synthetic data.

## Contribution

It introduces formal definitions of stereotyping as representational harms and explores their effects within the ML pipeline, along with mitigation methods.

## Key findings

- Formalization of stereotyping as representational harm
- Demonstration of stereotyping's impact on allocation harms
- Effective mitigation strategies on synthetic datasets

## Abstract

While harms of allocation have been increasingly studied as part of the subfield of algorithmic fairness, harms of representation have received considerably less attention. In this paper, we formalize two notions of stereotyping and show how they manifest in later allocative harms within the machine learning pipeline. We also propose mitigation strategies and demonstrate their effectiveness on synthetic datasets.