Optimizing generalized Gini indices for fairness in rankings

TL;DR

This paper introduces a novel approach using generalized Gini welfare functions to enhance fairness in ranking systems, effectively balancing item exposure and user satisfaction across diverse recommendation scenarios.

Contribution

It proposes a new fairness optimization framework based on GGFs for ranking, addressing non-differentiability with advanced optimization techniques, and demonstrates improved trade-offs in real datasets.

Findings

Better fairness-performance trade-offs than baselines

Effective optimization of non-differentiable fairness criteria

Scalable approach for large datasets with up to 15k users and items

Abstract

There is growing interest in designing recommender systems that aim at being fair towards item producers or their least satisfied users. Inspired by the domain of inequality measurement in economics, this paper explores the use of generalized Gini welfare functions (GGFs) as a means to specify the normative criterion that recommender systems should optimize for. GGFs weight individuals depending on their ranks in the population, giving more weight to worse-off individuals to promote equality. Depending on these weights, GGFs minimize the Gini index of item exposure to promote equality between items, or focus on the performance on specific quantiles of least satisfied users. GGFs for ranking are challenging to optimize because they are non-differentiable. We resolve this challenge by leveraging tools from non-smooth optimization and projection operators used in differentiable sorting. We present experiments using real datasets with up to 15k users and items, which show that our approach obtains better trade-offs than the baselines on a variety of recommendation tasks and fairness criteria.