Online certification of preference-based fairness for personalized recommender systems
This addresses fairness concerns in personalized recommender systems by moving beyond group-level parity to individual preferences, though it is incremental in applying bandit methods to a new audit problem.
The paper tackled the problem of auditing recommender systems for fairness by proposing envy-freeness as a granular criterion based on individual preferences, and developed a sample-efficient bandit algorithm with theoretical guarantees that does not harm user experience, achieving trade-offs on real-world datasets.
Recommender systems are facing scrutiny because of their growing impact on the opportunities we have access to. Current audits for fairness are limited to coarse-grained parity assessments at the level of sensitive groups. We propose to audit for envy-freeness, a more granular criterion aligned with individual preferences: every user should prefer their recommendations to those of other users. Since auditing for envy requires to estimate the preferences of users beyond their existing recommendations, we cast the audit as a new pure exploration problem in multi-armed bandits. We propose a sample-efficient algorithm with theoretical guarantees that it does not deteriorate user experience. We also study the trade-offs achieved on real-world recommendation datasets.