{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T02:10:04Z","timestamp":1750299004442,"version":"3.41.0"},"reference-count":40,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2025,5,30]],"date-time":"2025-05-30T00:00:00Z","timestamp":1748563200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62302187"],"award-info":[{"award-number":["62302187"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Hubei Provincial Natural Science Foundation of China","award":["2024AFB047"],"award-info":[{"award-number":["2024AFB047"]}]},{"name":"Hong Kong RGC","award":["C7004-22G (CRF), C5032-23G (CRF), CRS_PolyU501\/23 (CRS) and T43-513\/23-N (TRS)"],"award-info":[{"award-number":["C7004-22G (CRF), C5032-23G (CRF), CRS_PolyU501\/23 (CRS) and T43-513\/23-N (TRS)"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Model. Perform. Eval. Comput. Syst."],"published-print":{"date-parts":[[2025,6,30]]},"abstract":"<jats:p>Federated learning is a new learning paradigm which utilizes crowdsourced data stored at dispersed user devices (aka clients) to learn a global model. Studies have shown that even though data are kept on local devices, an adversary is still able to infer client information during the training process or from the learned model. Differential privacy has recently been introduced to deep learning model training, to protect data privacy of clients. Nonetheless, it exacerbates unfairness with the learned model among participating clients due to its uniform clipping and noise addition, even when the training loss function explicitly considers unfairness. To validate the impact of the differential privacy mechanism in federated learning, we carefully approximate the correlation between fairness performance across clients and the fundamental operations within the differential privacy mechanism and quantify the influence of differential privacy mechanisms on model performance across various clients. Subsequently, leveraging our theoretical findings regarding the effect of the differential privacy mechanism, we formulate the unfairness mitigation problem and propose an algorithm based on the modified method of differential multipliers. Extensive evaluation shows that our method outperforms state-of-the-art differentially private federated learning algorithm by about 30% for non-i.i.d. data distribution in terms of the variance of model performance across clients.<\/jats:p>","DOI":"10.1145\/3725847","type":"journal-article","created":{"date-parts":[[2025,3,25]],"date-time":"2025-03-25T11:56:36Z","timestamp":1742903796000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Mitigating Unfairness in Differentially-Private Federated Learning"],"prefix":"10.1145","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4825-8153","authenticated-orcid":false,"given":"Bingqian","family":"Du","sequence":"first","affiliation":[{"name":"Huazhong University of Science and Technology, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0165-4930","authenticated-orcid":false,"given":"Liyao","family":"Xiang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3144-4398","authenticated-orcid":false,"given":"Chuan","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of Hong Kong, Hong Kong, Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,5,30]]},"reference":[{"key":"e_1_3_2_2_2","first-page":"265","volume-title":"Proceedings of the 12th  \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Symposium on Operating Systems Design and Implementation ( \\(\\lbrace\\) OSDI \\(\\rbrace\\) \u201916)","author":"Abadi Mart\u00edn","year":"2016","unstructured":"Mart\u00edn Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et\u00a0al. 2016. Tensorflow: A system for large-scale machine learning. In Proceedings of the 12th \\(\\lbrace\\) USENIX \\(\\rbrace\\) Symposium on Operating Systems Design and Implementation ( \\(\\lbrace\\) OSDI \\(\\rbrace\\) \u201916). 265\u2013283."},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978318"},{"key":"e_1_3_2_4_2","first-page":"15479","volume-title":"Proceedings of the Advances in Neural Information Processing Systems","author":"Bagdasaryan Eugene","year":"2019","unstructured":"Eugene Bagdasaryan, Omid Poursaeed, and Vitaly Shmatikov. 2019. Differential privacy has disparate impact on model accuracy. In Proceedings of the Advances in Neural Information Processing Systems. 15479\u201315488."},{"key":"e_1_3_2_5_2","first-page":"394","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Balle Borja","year":"2018","unstructured":"Borja Balle and Yu-Xiang Wang. 2018. Improving the gaussian mechanism for differential privacy: Analytical calibration and optimal denoising. In Proceedings of the International Conference on Machine Learning. PMLR, 394\u2013403."},{"key":"e_1_3_2_6_2","unstructured":"Abhishek Bhowmick John Duchi Julien Freudiger Gaurav Kapoor and Ryan Rogers. 2018. Protection against reconstruction and its applications in private federated learning. arXiv:1812.00984. Retrieved from https:\/\/arxiv.org\/abs\/1812.00984"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2018.01.007"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1145\/3314183.3323847"},{"key":"e_1_3_2_9_2","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1137\/1.9781611976700.21","volume-title":"Proceedings of the 2021 SIAM International Conference on Data Mining","author":"Du Wei","year":"2021","unstructured":"Wei Du, Depeng Xu, Xintao Wu, and Hanghang Tong. 2021. Fairness-aware agnostic federated learning. In Proceedings of the 2021 SIAM International Conference on Data Mining. SIAM, 181\u2013189."},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/2090236.2090255"},{"key":"e_1_3_2_11_2","first-page":"7494","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Ezzeldin Yahya H.","year":"2023","unstructured":"Yahya H. Ezzeldin, Shen Yan, Chaoyang He, Emilio Ferrara, and A. Salman Avestimehr. 2023. Fairfed: Enabling group fairness in federated learning. In Proceedings of the AAAI Conference on Artificial Intelligence. 7494\u20137502."},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/3411501.3419419"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1145\/2810103.2813677"},{"key":"e_1_3_2_14_2","first-page":"16937","article-title":"Inverting gradients-how easy is it to break privacy in federated learning?","volume":"33","author":"Geiping Jonas","year":"2020","unstructured":"Jonas Geiping, Hartmut Bauermeister, Hannah Dr\u00f6ge, and Michael Moeller. 2020. Inverting gradients-how easy is it to break privacy in federated learning? Advances in Neural Information Processing Systems 33 (2020), 16937\u201316947.","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"12","key":"e_1_3_2_15_2","first-page":"2009","article-title":"Twitter sentiment classification using distant supervision","volume":"1","author":"Go Alec","year":"2009","unstructured":"Alec Go, Richa Bhayani, and Lei Huang. 2009. Twitter sentiment classification using distant supervision. CS224N Project Report, Stanford 1, 12 (2009), 2009.","journal-title":"CS224N Project Report, Stanford"},{"key":"e_1_3_2_16_2","first-page":"419","volume-title":"Proceedings of the International Conference on Network and System Security","author":"Gu Xiuting","year":"2020","unstructured":"Xiuting Gu, Tianqing Zhu, Jie Li, Tao Zhang, and Wei Ren. 2020. The impact of differential privacy on model fairness in federated learning. In Proceedings of the International Conference on Network and System Security. Springer, 419\u2013430."},{"key":"e_1_3_2_17_2","unstructured":"Andrew Hard Kanishka Rao Rajiv Mathews Swaroop Ramaswamy Fran\u00e7oise Beaufays Sean Augenstein Hubert Eichner Chlo\u00e9 Kiddon and Daniel Ramage. 2018. Federated learning for mobile keyboard prediction. arXiv preprint arXiv:1811.03604 (2018)."},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1145\/3133956.3134012"},{"key":"e_1_3_2_19_2","article-title":"Personalized federated learning with differential privacy","author":"Hu Rui","year":"2020","unstructured":"Rui Hu, Yuanxiong Guo, Hongning Li, Qingqi Pei, and Yanmin Gong. 2020. Personalized federated learning with differential privacy. IEEE Internet of Things Journal 7, 10 (2020), 9530\u20139539.","journal-title":"IEEE Internet of Things Journal"},{"key":"e_1_3_2_20_2","first-page":"3000","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Jagielski Matthew","year":"2019","unstructured":"Matthew Jagielski, Michael Kearns, Jieming Mao, Alina Oprea, Aaron Roth, Saeed Sharifi-Malvajerdi, and Jonathan Ullman. 2019. Differentially private fair learning. In Proceedings of the International Conference on Machine Learning. 3000\u20133008."},{"key":"e_1_3_2_21_2","unstructured":"Jakub Kone\u010dn\u1ef3 H. Brendan McMahan Felix X. Yu Peter Richt\u00e1rik Ananda Theertha Suresh and Dave Bacon. 2016. Federated learning: Strategies for improving communication efficiency. arXiv:1610.05492. Retrieved from https:\/\/arxiv.org\/abs\/1610.05492"},{"key":"e_1_3_2_22_2","unstructured":"Alex Krizhevsky Geoffrey Hinton and others. 2009. Learning multiple layers of features from tiny images. (2009). Retrieved from https:\/\/www.cs.utoronto.ca\/kriz\/learning-features-2009-%0ATR.pdf"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2975749"},{"key":"e_1_3_2_24_2","volume-title":"Proceedings of the International Conference on Learning Representations","author":"Li Tian","year":"2019","unstructured":"Tian Li, Maziar Sanjabi, Ahmad Beirami, and Virginia Smith. 2019. Fair resource allocation in federated learning. In Proceedings of the International Conference on Learning Representations."},{"key":"e_1_3_2_25_2","unstructured":"Xinpeng Ling Jie Fu Zhili Chen Kuncan Wang Huifa Li Tong Cheng Guanying Xu and Qin Li. 2024. FedFDP: Federated learning with fairness and differential privacy. arXiv:2402.16028. Retrieved from https:\/\/arxiv.org\/abs\/2402.16028"},{"key":"e_1_3_2_26_2","article-title":"How to democratise and protect AI: Fair and differentially private decentralised deep learning","author":"Lyu Lingjuan","year":"2022","unstructured":"Lingjuan Lyu, Yitong Li, Karthik Nandakumar, Jiangshan Yu, and Xingjun Ma. 2022. How to democratise and protect AI: Fair and differentially private decentralised deep learning. IEEE Transactions on Dependable and Secure Computing 19, Mar.-Apr. (2022), 1003\u20131017.","journal-title":"IEEE Transactions on Dependable and Secure Computing"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/MNET.001.1900506"},{"key":"e_1_3_2_28_2","first-page":"1273","volume-title":"Proceedings of the Artificial Intelligence and Statistics","author":"McMahan Brendan","year":"2017","unstructured":"Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. In Proceedings of the Artificial Intelligence and Statistics. 1273\u20131282."},{"key":"e_1_3_2_29_2","unstructured":"H. B. McMahan D. Ramage K. Talwar et\u00a0al. 2018. Learning differentially private recurrent language models. International Conference on Learning Representations. 2018."},{"key":"e_1_3_2_30_2","first-page":"4615","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Mohri Mehryar","year":"2019","unstructured":"Mehryar Mohri, Gary Sivek, and Ananda Theertha Suresh. 2019. Agnostic federated learning. In Proceedings of the International Conference on Machine Learning. PMLR, 4615\u20134625."},{"key":"e_1_3_2_31_2","first-page":"1000","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Shamir Ohad","year":"2014","unstructured":"Ohad Shamir, Nati Srebro, and Tong Zhang. 2014. Communication-efficient distributed optimization using an approximate newton-type method. In Proceedings of the International Conference on Machine Learning. PMLR, 1000\u20131008."},{"key":"e_1_3_2_32_2","first-page":"601","volume-title":"Proceedings of the 25th  \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\)  Security 16)","author":"Tram\u00e8r Florian","year":"2016","unstructured":"Florian Tram\u00e8r, Fan Zhang, Ari Juels, Michael K. Reiter, and Thomas Ristenpart. 2016. Stealing machine learning models via prediction apis. In Proceedings of the 25th \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security Symposium ( \\(\\lbrace\\) USENIX \\(\\rbrace\\) Security 16). 601\u2013618."},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1145\/3338501.3357370"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2020.2988575"},{"key":"e_1_3_2_35_2","unstructured":"Han Xiao Kashif Rasul and Roland Vollgraf. 2017. Fashion-mnist: A novel image dataset for benchmarking machine learning algorithms. arXiv:1708.07747. Retrieved from https:\/\/arxiv.org\/abs\/1708.07747"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2021.3073925"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1145\/3308560.3317584"},{"key":"e_1_3_2_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155359"},{"key":"e_1_3_2_39_2","first-page":"1051","volume-title":"Proceedings of the 2020 IEEE International Conference on Big Data","author":"Zhang Daniel Yue","year":"2020","unstructured":"Daniel Yue Zhang, Ziyi Kou, and Dong Wang. 2020. Fairfl: A fair federated learning approach to reducing demographic bias in privacy-sensitive classification models. In Proceedings of the 2020 IEEE International Conference on Big Data. IEEE, 1051\u20131060."},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMECH.2021.3065522"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2020.106679"}],"container-title":["ACM Transactions on Modeling and Performance Evaluation of Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3725847","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3725847","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:57:04Z","timestamp":1750298224000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3725847"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,30]]},"references-count":40,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,6,30]]}},"alternative-id":["10.1145\/3725847"],"URL":"https:\/\/doi.org\/10.1145\/3725847","relation":{},"ISSN":["2376-3639","2376-3647"],"issn-type":[{"type":"print","value":"2376-3639"},{"type":"electronic","value":"2376-3647"}],"subject":[],"published":{"date-parts":[[2025,5,30]]},"assertion":[{"value":"2024-04-25","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":"2025-05-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}