AI & softwareResearch findings
A Structure-Aware Fairness Attack Method for Recommender Systems
Original published 2026.06.28 04:50 KSTAdded to Blackdeck 2026.10.20 12:16 KST
What has changed?
Proposes a reinforcement learning-based method that considers both structural and sequential dependencies to worsen fairness
Key points
- Unfairness in recommender systems has raised significant social and ethical concerns
- Existing research has focused on performance degradation rather than fairness degradation
- The proposed method enhances unfairness by modeling structural and sequential dependencies in fake interactions
Connect it to your work
Models structural and sequential patterns of fake user-item interactions to amplify bias in recommendation systems
What remains unproven
This research note is based on an abstract; check the original for test conditions and application scope.
Originals used for this note
Links open the original publisher in a new tab.