Fairness in Recommendation Ranking through Pairwise Comparisons
Open accessThe DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index.
Abstract
Recommender systems are one of the most pervasive applications of machine learning in industry, with many services using them to match users to products or information. As such it is important to ask: what are the possible fairness risks, how can we quantify them, and how should we address them? In this paper we offer a set of novel metrics for evaluating algorithmic fairness concerns in recommender systems. In particular we show how measuring fairness based on pairwise comparisons from randomized experiments provides a tractable means to reason about fairness in rankings from recommender systems. Building on this metric, we offer a new regularizer to encourage improving this metric during model training and thus improve fairness in the resulting rankings. We apply this pairwise regularization to a large-scale, production recommender system and show that we are able to significantly improve the system's pairwise fairness.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Matrix Factorization Techniques for Recommender Systems
Koren, Y., Bell, R., & Volinsky, C. (2009) · Computer · doi:10.1109/mc.2009.263 · Cited by 11,673
Neural Collaborative Filtering
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S. (2017) · Proceedings of the 26th International Conference on World Wide Web · doi:10.1145/3038912.3052569 · Cited by 6,652
Optimizing search engines using clickthrough data
Joachims, T. (2002) · Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/775047.775067 · Cited by 3,873
Deep Neural Networks for YouTube Recommendations
Covington, P., Adams, J., & Sargin, E. (2016) · Proceedings of the 10th ACM Conference on Recommender Systems · doi:10.1145/2959100.2959190 · Cited by 3,539
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501
Exposure to ideologically diverse news and opinion on Facebook
Bakshy, E., Messing, S., & Adamic, L. A. (2015) · Science · doi:10.1126/science.aaa1160 · Cited by 3,245
Equality of Opportunity in Supervised Learning
Hardt, M., Price, E., & Srebro, N. (2016) · arXiv · doi:10.48550/arxiv.1610.02413 · Cited by 2,536 · cited 6×
“[23], where we are concerned with differences in accuracy across groups.”
BackgroundInfluentialLearning to rank: from pairwise approach to listwise approach
Cao, Z., Qin, T., Liu, T.-Y., Tsai, M.-F., & Li, H. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273513 · Cited by 1,960 · cited 2×
“(This analysis is similar to probabilistic interpretations in traditional pairwise IR [14], but now in the context of recommender system fairness.”
BackgroundMitigating Unwanted Biases with Adversarial Learning
Zhang, B. H., Lemoine, B., & Mitchell, M. (2018) · Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3278721.3278779 · Cited by 1,387 · cited 1×
“wn for demographic data. Rather, numerous approaches have been developed for optimizing fairness metrics during classier training, such as constraint-based optimization [2, 22], adversarial learning [53, 35, 20, 7, 37, 54], and regularization over model predictions [30, 51, 5, 10]. We build on these regularization app…”
BackgroundA Reductions Approach to Fair Classification
Agarwal, A., Beygelzimer, A., Dudík, M., Langford, J., & Wallach, H. (2018) · arXiv · doi:10.48550/arxiv.1803.02453 · Cited by 1,288
Cited by
A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions
Gao, C., Zheng, Y., Li, N., Li, Y., Qin, Y., Piao, J., Quan, Y., Chang, J., Jin, D., He, X., et al. (2023) · ACM Transactions on Recommender Systems · doi:10.1145/3568022 · Cited by 1,082
Bias and Debias in Recommender System: A Survey and Future Directions
Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023) · ACM Transactions on Information Systems · doi:10.1145/3564284 · Cited by 741
Self-citationA Review on Fairness in Machine Learning
Pessach, D., & Shmueli, E. (2023) · ACM Computing Surveys · doi:10.1145/3494672 · Cited by 598
“Beutel et al. [21] propose several measures based on pairwise fairness .”
BackgroundFairness in machine learning: A survey
Caton, S., & Haas, C. (2024) · ACM Computing Surveys · doi:10.1145/3616865 · Cited by 646
A Survey on the Fairness of Recommender Systems
Wang, Y., Ma, W., Zhang, M., Liu, Y., & Ma, S. (2023) · ACM Transactions on Information Systems · doi:10.1145/3547333 · Cited by 356
Multistakeholder recommendation: Survey and research directions
Abdollahpouri, H., Adomavicius, G., Burke, R., Guy, I., Jannach, D., Kamishima, T., Krasnodebski, J., & Pizzato, L. (2020) · User Modeling and User-Adapted Interaction · doi:10.1007/s11257-019-09256-1 · Cited by 319
Evaluating Recommender Systems: Survey and Framework
Zangerle, E., & Bauer, C. (2023) · ACM Computing Surveys · doi:10.1145/3556536 · Cited by 283
Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation
Zhang, J., Bao, K., Zhang, Y., Wang, W., Feng, F., & He, X. (2023) · arXiv · doi:10.48550/arxiv.2305.07609 · Cited by 196
An empirical characterization of fair machine learning for clinical risk prediction
Pfohl, S. R., Foryciarz, A., & Shah, N. H. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2020.103621 · Cited by 164
Reducing disparate exposure in ranking: A learning to rank approach
Zehlike, M., & Castillo, C. (2020) · Proceedings of The Web Conference 2020 · doi:10.1145/3366424.3380048 · Cited by 159
“[2] cares about disparate treatment, our listwise method directly optimizes the actual exposure a protected group would get, and is hence concerned with disparate impact.”
MethodFairness in Machine Learning: A Survey
Caton, S., & Haas, C. (2020) · arXiv · doi:10.48550/arxiv.2010.04053 · Cited by 646
Fairness and Bias in Algorithmic Hiring: A Multidisciplinary Survey
Fabris, A., Baranowska, N., Dennis, M. J., Graus, D., Hacker, P., Saldivar, J., Zuiderveen Borgesius, F., & Biega, A. J. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3696457 · Cited by 117
In-Processing Modeling Techniques for Machine Learning Fairness: A Survey
Wan, M., Zha, D., Liu, N., & Zou, N. (2023) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3551390 · Cited by 99
“Other notable regularizers include minimizing Wasserstein-1 distances between the classiier outputs and sensitive information [43], absolute correlation regularization that takes a simpliied view by minimizing the absolute value of the correlation between the sensitive groups and the predictions [11], and residual bet…”
BackgroundBalayn, A., Lofi, C., & Houben, G.-J. (2021) · The VLDB Journal · doi:10.1007/s00778-021-00671-8 · Cited by 98
“Works stemming from the machine learning and data mining communities have started to tackle unfairness from certain angles like evaluating the outputs of trained models [211]; and mitigating unfairness by post-processing the outputs of the system [33,40,70], or modifying the training process of the inference algorithm…”
BackgroundAddressing Marketing Bias in Product Recommendations
Wan, M., Ni, J., Misra, R., & McAuley, J. (2020) · Proceedings of the 13th International Conference on Web Search and Data Mining · doi:10.1145/3336191.3371855 · Cited by 81
“Following previous work using the regularizing schemes [1, 3, 30], we propose a fairness-aware framework by considering an error correlation loss to regularize systematic error biases on the market:”
BackgroundMethodInfluentialThe Sociodemographic Biases in Machine Learning Algorithms: A Biomedical Informatics Perspective
Franklin, G., Stephens, R., Piracha, M., Tiosano, S., Lehouillier, F., Koppel, R., & Elkin, P. (2024) · Life · doi:10.3390/life14060652 · Cited by 76
“…through algorithmic modifications that look at machine learning models’ loss of function by integrating “regularization” language specifically for the presumption of fairness, “for clas-sification” [91,92], “for ranking” [93], “for matching tasks” [94,95], and “for recommender systems” [96].”
BackgroundFairness metrics and bias mitigation strategies for rating predictions
Ashokan, A., & Haas, C. (2021) · Information Processing & Management · doi:10.1016/j.ipm.2021.102646 · Cited by 47
Fairness-aware Federated Matrix Factorization
Liu, S., Ge, Y., Xu, S., Zhang, Y., & Marian, A. (2022) · Proceedings of the 16th ACM Conference on Recommender Systems · doi:10.1145/3523227.3546771 · Cited by 43
“The in-processing methods aim at encoding fairness as part of the objective function, typically as a regularizer, and mitigate the bias during training [1, 4].”
MethodProvider fairness across continents in collaborative recommender systems
Gómez, E., Boratto, L., & Salamó, M. (2022) · Information Processing & Management · doi:10.1016/j.ipm.2021.102719 · Cited by 39
Pfohl, S., Xu, Y., Foryciarz, A., Ignatiadis, N., Genkins, J., & Shah, N. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533166 · Cited by 30
FAIR : Fairness‐aware information retrieval evaluation
Gao, R., Ge, Y., & Shah, C. (2022) · Journal of the Association for Information Science and Technology · doi:10.1002/asi.24648 · Cited by 19
“…can be improved by re-ranking the items based on the fairness constraints (Gao & Shah, 2020), or by applying fairness as a regularizer in the objective function (Asudeh et al., 2019; Beutel et al., 2019; L. E. Celis et al., 2019; Mehrotra et al., 2018; Wan et al., 2020; Yao & Huang, 2017a).”
BackgroundBeyond Individual and Group Fairness
Awasthi, P., Cortes, C., Mansour, Y., & Mohri, M. (2020) · arXiv · doi:10.48550/arxiv.2008.09490 · Cited by 8
“There has also been work, however, studying fairness criteria and algorithms for ranking [Celis et al., 2018, Beutel et al., 2019, Narasimhan et al., 2019] and clustering problems [Chierichetti et al., 2017, Schmidt et al., 2018, Backurs et al., 2019] with requirement of a proportional…”
BackgroundA bias detection tree approach for detecting disparities in a recommendation model’s errors
Misztal-Radecka, J., & Indurkhya, B. (2023) · User Modeling and User-Adapted Interaction · doi:10.1007/s11257-022-09334-x · Cited by 4
“However, to the best of our knowledge, all of the existing bias detection approaches also require that the attributes are known (Sánchez and Bellogín 2019; Beutel et al. 2017, 2019; Wan et al. 2020; Yao and Huang 2017; Kershaw et al. 2021; Wei et al. 2021; Zhu et al. 2021).”
BackgroundMethodResultInfluentialLocal Law 144: A Critical Analysis of Regression Metrics
Filippi, G., Zannone, S., Hilliard, A., & Koshiyama, A. (2023) · arXiv · doi:10.48550/arxiv.2302.04119 · Cited by 3
Metadata
- Authors
- Beutel, A., Chen, J., Doshi, T., Qian, H., Wei, L., Wu, Y., Heldt, L., Zhao, Z., Hong, L., Chi, E. H., & Goodrow, C.
- Venue
- Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
- Publisher
- ACM
- Published
- 2019-07-25
- Pages
- 2212-2220
- OpenAlex
- W2920187767
- Cited by
- 362
- References
- 57
- FWCI
- 46.66
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.