A Cost-Sensitive Meta-Learning Strategy for Fair Provider Exposure in Recommendation

TL;DR

The paper addresses the problem of fair provider exposure in recommender systems by ensuring minority provider groups receive exposure proportional to their representation in the catalog. It introduces a cost-sensitive meta-learning sampling strategy that modulates the triplet construction distribution via a parameter $C$ and a vector $p$ to balance positive and negative items across provider groups. Empirical results on MovieLens-1M and COCO demonstrate that the method preserves ranking utility (NDCG) while adjusting exposure to closely reflect catalog proportions, reducing bias without sacrificing performance. The approach is adaptable to other pairwise methods and datasets, and the authors provide public code to support reproducibility and further exploration of fairness in recommendations.

Abstract

When devising recommendation services, it is important to account for the interests of all content providers, encompassing not only newcomers but also minority demographic groups. In various instances, certain provider groups find themselves underrepresented in the item catalog, a situation that can influence recommendation results. Hence, platform owners often seek to regulate the exposure of these provider groups in the recommended lists. In this paper, we propose a novel cost-sensitive approach designed to guarantee these target exposure levels in pairwise recommendation models. This approach quantifies, and consequently mitigate, the discrepancies between the volume of recommendations allocated to groups and their contribution in the item catalog, under the principle of equity. Our results show that this approach, while aligning groups exposure with their assigned levels, does not compromise to the original recommendation utility. Source code and pre-processed data can be retrieved at https://github.com/alessandraperniciano/meta-learning-strategy-fair-provider-exposure.