Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Achieving Fairness via Post-Processing in Web-Scale Recommender Systems
(Submitted on 19 Jun 2020 (v1), last revised 11 Aug 2022 (this version, v3))
Abstract: Building fair recommender systems is a challenging and crucial area of study due to its immense impact on society. We extended the definitions of two commonly accepted notions of fairness to recommender systems, namely equality of opportunity and equalized odds. These fairness measures ensure that equally "qualified" (or "unqualified") candidates are treated equally regardless of their protected attribute status (such as gender or race). We propose scalable methods for achieving equality of opportunity and equalized odds in rankings in the presence of position bias, which commonly plagues data generated from recommender systems. Our algorithms are model agnostic in the sense that they depend only on the final scores provided by a model, making them easily applicable to virtually all web-scale recommender systems. We conduct extensive simulations as well as real-world experiments to show the efficacy of our approach.
Submission history
From: Preetam Nandy [view email][v1] Fri, 19 Jun 2020 20:12:13 GMT (172kb,D)
[v2] Fri, 5 Feb 2021 17:30:14 GMT (187kb,D)
[v3] Thu, 11 Aug 2022 06:42:18 GMT (4649kb,D)
Link back to: arXiv, form interface, contact.