Reproducibility Study: Joint Multisided Exposure Fairness for Recommendation

As part of the ML Reproducibility Challenge 2022 (MLRC), we conducted an independent reproducibility study of Joint Multisided Exposure Fairness for Recommendation — a method addressing exposure fairness for both users and items in recommender systems.

Paper: openreview.net/forum?id=A0Sjs3IJWb- Venue: ReScience / ML Reproducibility Challenge 2023 Authors: Hu, A., Ranum, O., Pozrikidou, C., Zhou, M.

Summary: We reproduced the key claims of the original paper, including convergence behaviour, fairness metric improvements, and baseline comparisons across multiple datasets. Our study confirms the central findings while documenting implementation details and hyperparameter sensitivity not covered in the original work.