Image search and retrieval tasks can perpetuate harmful stereotypes, erase cultural identities, and amplify social disparities. Current approaches to mitigate these representational harms balance the number of retrieved items across population groups defined by a small number of (often binary) attributes. However, most existing methods overlook intersectional groups determined by combinations of group attributes, such as gender, race, and ethnicity. We introduce Multi-Group Proportional Representation (MPR), a novel metric that measures representation across intersectional groups. We develop practical methods for estimating MPR, provide theoretical guarantees, and propose optimization algorithms to ensure MPR in retrieval. We demonstrate that existing methods optimizing for equal and proportional representation metrics may fail to promote MPR. Crucially, our work shows that optimizing MPR yields more proportional representation across multiple intersectional groups specified by a rich function class, often with minimal compromise in retrieval accuracy.
@article{arxiv.2407.08571,
title = {Multi-Group Proportional Representation in Retrieval},
author = {Alex Oesterling and Claudio Mayrink Verdun and Carol Xuan Long and Alexander Glynn and Lucas Monteiro Paes and Sajani Vithana and Martina Cardone and Flavio P. Calmon},
journal= {arXiv preprint arXiv:2407.08571},
year = {2024}
}
Comments
48 pages, 33 figures. Accepted as poster at NeurIPS 2024. Code can be found at https://github.com/alex-oesterling/multigroup-proportional-representation