Stairway to Fairness: Connecting Group and Individual Fairness

2025-08-29

AI Summary

The interplay between group fairness and individual fairness in recommender systems lacks a unified evaluation foundation, hindering systematic analysis of their mutual influence. Method: We propose the first unified evaluation framework, conducting systematic comparisons across three public datasets and eight state-of-the-art recommendation models. Contribution/Results: Our empirical study is the first to reveal an inherent trade-off: improving group fairness frequently degrades individual fairness. Specifically, recommendations achieving high group fairness often induce substantial individual-level unfairness—challenging the widely held implicit assumption that fairness objectives are jointly optimizable. These findings provide a novel theoretical perspective and practical caution for fair recommendation design, highlighting the need for explicit multi-faceted fairness considerations. To ensure transparency and reproducibility, we publicly release our implementation.

Technology Category

• Machine Learning: Ethics, Bias, and Fairness
• Data Mining & Knowledge Management: Recommender Systems
• Game Theory and Economic Paradigms: Fair Division

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Fairness in recommender systems (RSs) is commonly categorised into group fairness and individual fairness. However, there is no established scientific understanding of the relationship between the two fairness types, as prior work on both types has used different evaluation measures or evaluation objectives for each fairness type, thereby not allowing for a proper comparison of the two. As a result, it is currently not known how increasing one type of fairness may affect the other. To fill this gap, we study the relationship of group and individual fairness through a comprehensive comparison of evaluation measures that can be used for both fairness types. Our experiments with 8 runs across 3 datasets show that recommendations that are highly fair for groups can be very unfair for individuals. Our finding is novel and useful for RS practitioners aiming to improve the fairness of their systems. Our code is available at: https://github.com/theresiavr/stairway-to-fairness.

Problem

• Investigates relationship between group and individual fairness
• Examines how improving one fairness type affects the other
• Compares evaluation measures for both fairness types

Innovation

• Compares group and individual fairness measures
• Reveals trade-offs between fairness types
• Provides code for fairness evaluation