Fairness Through Regularization for Learning to Rank
arXiv:2102.05996
Abstract
Given the abundance of applications of ranking in recent years, addressing fairness concerns around automated ranking systems becomes necessary for increasing the trust among end-users. Previous work on fair ranking has mostly focused on application-specific fairness notions, often tailored to online advertising, and it rarely considers learning as part of the process. In this work, we show how to transfer numerous fairness notions from binary classification to a learning to rank setting. Our formalism allows us to design methods for incorporating fairness objectives with provable generalization guarantees. An extensive experimental evaluation shows that our method can improve ranking fairness substantially with no or only little loss of model quality.
arXiv comment34 pages
References in corpus (23)
- Equality of Opportunity in Supervised Learning
- Counterfactual Fairness
- Fairness of Exposure in Rankings
- FA*IR: A Fair Top-k Ranking Algorithm
- Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search
- Passage Re-ranking with BERT
- Controlling Fairness and Bias in Dynamic Learning-to-Rank
- Learning Non-Discriminatory Predictors
- Designing Fair Ranking Schemes
- Multisided Fairness for Recommendation
- Learning-to-Rank with BERT in TF-Ranking
- Policy Learning for Fairness in Ranking
- Fairness for Robust Log Loss Classification
- The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metric
- A Fairness-aware Hybrid Recommender System
- Fairness risk measures
- FACT: A Diagnostic for Group Fairness Trade-offs
- Learning Fair Representations for Kernel Models
- Rényi Fair Inference
- Learning Fair Scoring Functions: Bipartite Ranking under ROC-based Fairness Constraints
- On the Problem of Underranking in Group-Fair Ranking
- Fair Division Without Disparate Impact
- Individually Fair Ranking