{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T05:30:26Z","timestamp":1782279026252,"version":"3.54.5"},"reference-count":44,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T00:00:00Z","timestamp":1777507200000},"content-version":"vor","delay-in-days":275,"URL":"http:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100000006","name":"Office of Naval Research","doi-asserted-by":"crossref","award":["N00014-23-1-2590"],"award-info":[{"award-number":["N00014-23-1-2590"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"crossref"}]},{"name":"National Science Foundation","award":["2231174, 2310831, 2428059, 2435696, 2440954"],"award-info":[{"award-number":["2231174, 2310831, 2428059, 2435696, 2440954"]}]},{"name":"Michigan Institute for Data Science Propelling Original Data Science"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Recomm. Syst."],"published-print":{"date-parts":[[2026,3,31]]},"abstract":"<jats:p>\n            Recommender systems play a key role in shaping modern web ecosystems. These systems alternate between (1) making recommendations, (2) collecting user responses to these recommendations, and (3) retraining the recommendation algorithm based on this feedback. During this process, the recommender influences the user behavioral data that is subsequently used to update the recommender itself, thus creating a feedback loop. Recent work has shown that feedback loops may compromise recommendation quality and homogenize user behavior, raising ethical and performance concerns around deploying recommender systems. To address these concerns, we propose the\n            <jats:italic toggle=\"yes\">causal adjustment for feedback loops (CAFL)<\/jats:italic>\n            , an algorithm that uses causal inference to break feedback loops for any loss-minimizing recommendation algorithms. The key observation is that a recommender system does not suffer from feedback loops if it reasons about causal quantities, namely the intervention distributions of recommendations on user ratings. Moreover, we can calculate these intervention distributions from observational data by adjusting for the recommender system\u2019s predictions of user preferences. Using simulated environments, we demonstrate that CAFL improves recommendation quality when compared to prior correction methods.\n          <\/jats:p>","DOI":"10.1145\/3728372","type":"journal-article","created":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T07:23:59Z","timestamp":1745997839000},"page":"1-20","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Breaking Feedback Loops in Recommender Systems with Causal Inference"],"prefix":"10.1145","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9977-9813","authenticated-orcid":false,"given":"Karl","family":"Krauth","sequence":"first","affiliation":[{"name":"Stanford University","place":["Stanford, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6617-4842","authenticated-orcid":false,"given":"Yixin","family":"Wang","sequence":"additional","affiliation":[{"name":"University of Michigan","place":["Ann Arbor, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8935-817X","authenticated-orcid":false,"given":"Michael","family":"Jordan","sequence":"additional","affiliation":[{"name":"UC Berkeley","place":["Berkeley, United States"]},{"name":"INRIA","place":["Berkeley, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,7,29]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2007.90"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1145\/3240323.3240360"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1080\/15427951.2006.10129129"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/3240323.3240370"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1145\/3460231.3474274"},{"key":"e_1_3_2_7_2","unstructured":"Sebastian Farquhar Yarin Gal and Tom Rainforth. 2021. On statistical bias in active learning: How and when to fix it. International Conference on Learning Representations (2021)."},{"key":"e_1_3_2_8_2","first-page":"326","volume-title":"Proceedings of the UAI","author":"Gopalan Prem","year":"2015","unstructured":"Prem Gopalan, Jake M. Hofman, and David M. Blei. 2015. Scalable recommendation with hierarchical poisson factorization. In Proceedings of the UAI. 326\u2013335."},{"key":"e_1_3_2_9_2","first-page":"3176","volume-title":"Proceedings of the Advances in Neural Information Processing Systems 27.","author":"Gopalan Prem K.","year":"2014","unstructured":"Prem K. Gopalan, Laurent Charlin, and David Blei. 2014. Content-based recommendations with poisson factorization. In Proceedings of the Advances in Neural Information Processing Systems 27.Z. Ghahramani, M. Welling, C. Cortes, N.D. Lawrence, and K.Q. Weinberger (Eds.), Curran Associates, Inc., 3176\u20133184."},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/2827872"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.1145\/3038912.3052569"},{"key":"e_1_3_2_12_2","first-page":"1512","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Hern\u00e1ndez-Lobato Jos\u00e9 Miguel","year":"2014","unstructured":"Jos\u00e9 Miguel Hern\u00e1ndez-Lobato, Neil Houlsby, and Zoubin Ghahramani. 2014. Probabilistic matrix factorization with non-random missing data. In Proceedings of the International Conference on Machine Learning. PMLR, 1512\u20131520."},{"key":"e_1_3_2_13_2","unstructured":"Homa Hosseinmardi Amir Ghasemian Aaron Clauset David M Rothschild Markus Mobius and Duncan J Watts. 2020. Evaluating the scale growth and origins of right-wing echo chambers on YouTube. arXiv:2011.12843. Retrieved from https:\/\/arxiv.org\/abs\/2011.12843"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2008.22"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9781139025751"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1145\/3306618.3314288"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467298"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1145\/1557019.1557072"},{"key":"e_1_3_2_19_2","unstructured":"Karl Krauth Sarah Dean Alex Zhao Wenshuo Guo Mihaela Curmei Benjamin Recht and Michael I Jordan. 2020. Do Offline metrics predict online performance in recommender systems? arXiv:2011.07931. Retrieved from https:\/\/arxiv.org\/abs\/2011.07931"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.5555\/2946645.2946660"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1145\/3404835.3463118"},{"key":"e_1_3_2_22_2","doi-asserted-by":"publisher","DOI":"10.1145\/2872427.2883090"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1145\/3178876.3186150"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/3340531.3412152"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/1639714.1639717"},{"key":"e_1_3_2_26_2","first-page":"1257","volume-title":"Proceedings of the Advances in Neural Information Processing Systems","author":"Mnih Andriy","year":"2008","unstructured":"Andriy Mnih and Ruslan R. Salakhutdinov. 2008. Probabilistic matrix factorization. In Proceedings of the Advances in Neural Information Processing Systems. 1257\u20131264."},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2011.134"},{"key":"e_1_3_2_28_2","unstructured":"Weishen Pan Sen Cui Hongyi Wen Kun Chen Changshui Zhang and Fei Wang. 2021. Correcting the user feedback-loop bias for recommendation systems. arXiv:2109.06037. Retrieved from https:\/\/arxiv.org\/abs\/2109.06037"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511803161"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1145\/3539618.3592077"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/2168752.2168771"},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1145\/3336191.3371783"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1126\/science.1121066"},{"key":"e_1_3_2_34_2","first-page":"862","volume-title":"Proceedings of the International Conference on Artificial Intelligence and Statistics","author":"Schmit Sven","year":"2018","unstructured":"Sven Schmit and Carlos Riquelme. 2018. Human interaction with recommendation systems. In Proceedings of the International Conference on Artificial Intelligence and Statistics. PMLR, 862\u2013870."},{"key":"e_1_3_2_35_2","first-page":"1670","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Schnabel Tobias","year":"2016","unstructured":"Tobias Schnabel, Adith Swaminathan, Ashudeep Singh, Navin Chandak, and Thorsten Joachims. 2016. Recommendations as treatments: Debiasing learning and evaluation. In Proceedings of the International Conference on Machine Learning. PMLR, 1670\u20131679."},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1145\/2740908.2742726"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1145\/2764468.2764488"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.5555\/3157382.3157461"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308558.3313710"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308560.3317303"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1145\/1148170.1148257"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1145\/3383313.3412225"},{"key":"e_1_3_2_43_2","unstructured":"Yixin Wang Dhanya Sridhar and David Blei. 2023. Adjusting machine learning decisions for equal opportunity and counterfactual fairness. Transactions on Machine Learning Research (2023)."},{"key":"e_1_3_2_44_2","unstructured":"Hangtong Xu Yuanbo Xu Yongjian Yang Fuzhen Zhuang and Hui Xiong. 2023. DPR: An algorithm mitigate bias accumulation in recommendation feedback loops. arXiv:2311.05864. Retrieved from https:\/\/arxiv.org\/abs\/2311.05864"},{"key":"e_1_3_2_45_2","unstructured":"Yin Zheng Bangsheng Tang Wenkui Ding and Hanning Zhou. 2016. A neural autoregressive approach to collaborative filtering. In International Conference on Machine Learning. PMLR 764\u2013773."}],"container-title":["ACM Transactions on Recommender Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3728372","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3728372","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,29]],"date-time":"2025-07-29T12:47:19Z","timestamp":1753793239000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3728372"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,29]]},"references-count":44,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,3,31]]}},"alternative-id":["10.1145\/3728372"],"URL":"https:\/\/doi.org\/10.1145\/3728372","relation":{},"ISSN":["2770-6699"],"issn-type":[{"value":"2770-6699","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,29]]},"assertion":[{"value":"2024-08-12","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-07","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-29","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}