User Fairness, Item Fairness, and Diversity for Rankings in Two-Sided Markets

Abstract

Ranking items by their probability of relevance has long been the goal of conventional ranking systems. While this maximizes traditional criteria of ranking performance, there is a growing understanding that it is an oversimplification in online platforms that serve not only a diverse user population, but also the producers of the items. In particular, ranking algorithms are expected to be fair in how they serve all groups of users -- not just the majority group -- and they also need to be fair in how they divide exposure among the items. These fairness considerations can partially be met by adding diversity to the rankings, as done in several recent works. However, we show in this paper that user fairness, item fairness and diversity are fundamentally different concepts. In particular, we find that algorithms that consider only one of the three desiderata can fail to satisfy and even harm the other two. To overcome this shortcoming, we present the first ranking algorithm that explicitly enforces all three desiderata. The algorithm optimizes user and item fairness as a convex optimization problem which can be solved optimally. From its solution, a ranking policy can be derived via a novel Birkhoff-von Neumann decomposition algorithm that optimizes diversity. Beyond the theoretical analysis, we investigate empirically on a new benchmark dataset how effectively the proposed ranking algorithm can control user fairness, item fairness and diversity, as well as the trade-offs between them.

arXiv comment19 pages, 3 figures, ICTIR 2021

References in corpus (9)

• FA*IR: A Fair Top-k Ranking Algorithm
• Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search
• Equity of Attention: Amortizing Individual Fairness in Rankings
• Beyond Personalization: Research Directions in Multistakeholder Recommendation
• Designing Fair Ranking Schemes
• Beyond Parity: Fairness Objectives for Collaborative Filtering
• Fighting Fire with Fire: Using Antidote Data to Improve Polarization and Fairness of Recommender Systems
• Policy Learning for Fairness in Ranking
• The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metric

Cited by in corpus (14)

• Building Human Values into Recommender Systems: An Interdisciplinary Synthesis
• Fairness of Exposure in Light of Incomplete Exposure Estimation
• Understanding and Mitigating the Effect of Outliers in Fair Ranking
• Distributionally-Informed Recommender System Evaluation
• Introducing the Expohedron for Efficient Pareto-optimal Fairness-Utility Amortizations in Repeated Rankings
• Fairness of Exposure in Stochastic Bandits
• Predictive Uncertainty-based Bias Mitigation in Ranking
• Subgroup Fairness in Two-Sided Markets
• FairSort: Learning to Fair Rank for Personalized Recommendations in Two-Sided Platforms
• Group fairness without demographics using social networks
• Pareto-Optimal Fairness-Utility Amortizations in Rankings with a DBN Exposure Model
• Fairness in Ranking under Disparate Uncertainty
• Learning to Re-rank with Constrained Meta-Optimal Transport
• A Reproducibility Study of Product-side Fairness in Bundle Recommendation