{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T18:38:29Z","timestamp":1781375909416,"version":"3.54.1"},"reference-count":52,"publisher":"Association for Computing Machinery (ACM)","issue":"5","license":[{"start":{"date-parts":[[2022,6,11]],"date-time":"2022-06-11T00:00:00Z","timestamp":1654905600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Intell. Syst. Technol."],"published-print":{"date-parts":[[2022,10,31]]},"abstract":"\n Federated Learning (FL) [\n 31<\/jats:xref>\n ] is a decentralized learning mechanism that has attracted increasing attention due to its achievements in computational efficiency and privacy preservation. However, recent research highlights that the original FL framework may still reveal sensitive information of clients\u2019 local data from the exchanged local updates and the global model parameters. Local Differential Privacy (LDP), as a rigorous definition of privacy, has been applied to Federated Learning to provide formal privacy guarantees and prevent potential privacy leakage. However, previous LDP-FL solutions suffer from considerable utility loss with an increase of model dimensionality. Recent work [\n 29<\/jats:xref>\n ] proposed a two-stage framework that mitigates the dimension-dependency problem by first selecting one \u201cimportant\u201d dimension for each local update and then perturbing the dimension value to construct the sparse privatized update. However, the framework may still suffer from utility loss because of the insufficient per-stage privacy budget and slow model convergence.\n <\/jats:p>\n \n In this article, we propose an improved framework,\n SignDS-FL<\/jats:sc>\n , which shares the concept of dimension selection with Reference [\n 29<\/jats:xref>\n ], but saves the privacy cost for the value perturbation stage by assigning random sign values to the selected dimensions. Besides using the single-dimension selection algorithms in Reference [\n 29<\/jats:xref>\n ], we propose an Exponential Mechanism-based Multi-Dimension Selection algorithm that further improves model convergence and accuracy. We evaluate the framework on a number of real-world datasets with both simple logistic regression models and deep neural networks. For training logistic regression models on structured datasets, our framework yields only a\n \n \\( \\sim \\)<\/jats:tex-math>\n <\/jats:inline-formula>\n 1%\u20132% accuracy loss in comparison to a\n \n \\( \\sim \\)<\/jats:tex-math>\n <\/jats:inline-formula>\n 5%\u201315% decrease of accuracy for the baseline methods. For training deep neural networks on image datasets, the accuracy loss of our framework is less than\n \n \\( 8\\% \\)<\/jats:tex-math>\n <\/jats:inline-formula>\n and at best only\n \n \\( 2\\% \\)<\/jats:tex-math>\n <\/jats:inline-formula>\n . Extensive experimental results show that our framework significantly outperforms the previous LDP-FL solutions and enjoys an advanced utility-privacy balance.\n <\/jats:p>","DOI":"10.1145\/3517820","type":"journal-article","created":{"date-parts":[[2022,3,22]],"date-time":"2022-03-22T14:20:27Z","timestamp":1647958827000},"page":"1-22","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":19,"title":["SignDS-FL: Local Differentially Private Federated Learning with Sign-based Dimension Selection"],"prefix":"10.1145","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4959-9910","authenticated-orcid":false,"given":"Xue","family":"Jiang","sequence":"first","affiliation":[{"name":"Technical University of Munich, Garching, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuebing","family":"Zhou","sequence":"additional","affiliation":[{"name":"Huawei Technologies D\u00fcsseldorf GmbH, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jens","family":"Grossklags","sequence":"additional","affiliation":[{"name":"Technical University of Munich, Garching, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2022,6,11]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978318"},{"key":"e_1_3_2_3_2","volume-title":"Proceedings of the 21st European Symposium on Artificial Neural Networks","author":"Anguita Davide","year":"2013","unstructured":"Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra, and Jorge Luis Reyes-Ortiz. 2013. A public domain dataset for human activity recognition using smartphones. In Proceedings of the 21st European Symposium on Artificial Neural Networks. 437\u2013442."},{"key":"e_1_3_2_4_2","volume-title":"Proceedings of the 8th International Conference on Learning Representations","author":"Augenstein Sean","year":"2020","unstructured":"Sean Augenstein, H. Brendan McMahan, Daniel Ramage, Swaroop Ramaswamy, Peter Kairouz, Mingqing Chen, Rajiv Mathews, and Blaise Ag\u00fcera y Arcas. 2020. Generative models for effective ML on private, decentralized datasets. In Proceedings of the 8th International Conference on Learning Representations."},{"key":"e_1_3_2_5_2","first-page":"559","volume-title":"Proceedings of the 35th International Conference on Machine Learning","volume":"80","author":"Bernstein Jeremy","year":"2018","unstructured":"Jeremy Bernstein, Yu-Xiang Wang, Kamyar Azizzadenesheli, and Animashree Anandkumar. 2018. SIGNSGD: Compressed optimisation for non-convex problems. In Proceedings of the 35th International Conference on Machine Learning, Vol. 80. PMLR, 559\u2013568."},{"key":"e_1_3_2_6_2","article-title":"Protection against reconstruction and its applications in private federated learning","volume":"1812","author":"Bhowmick Abhishek","year":"2018","unstructured":"Abhishek Bhowmick, John C. Duchi, Julien Freudiger, Gaurav Kapoor, and Ryan Rogers. 2018. Protection against reconstruction and its applications in private federated learning. CoRR abs\/1812.00984.","journal-title":"CoRR"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOMWKSHPS51825.2021.9484526"},{"key":"e_1_3_2_8_2","article-title":"Expanding the reach of federated learning by reducing client resource requirements","volume":"1812","author":"Caldas Sebastian","year":"2018","unstructured":"Sebastian Caldas, Jakub Kone\u010dn\u00fd, H. Brendan McMahan, and Ameet Talwalkar. 2018. Expanding the reach of federated learning by reducing client resource requirements. CoRR abs\/1812.07210.","journal-title":"CoRR"},{"key":"e_1_3_2_9_2","first-page":"289","volume-title":"Proceedings of the 22nd Annual Conference on Neural Information Processing Systems","author":"Chaudhuri Kamalika","year":"2008","unstructured":"Kamalika Chaudhuri and Claire Monteleoni. 2008. Privacy-preserving logistic regression. In Proceedings of the 22nd Annual Conference on Neural Information Processing Systems. Curran Associates, Inc., 289\u2013296."},{"key":"e_1_3_2_10_2","article-title":"EMNIST: An extension of MNIST to handwritten letters","volume":"1702","author":"Cohen Gregory","year":"2017","unstructured":"Gregory Cohen, Saeed Afshar, Jonathan Tapson, and Andr\u00e9 van Schaik. 2017. EMNIST: An extension of MNIST to handwritten letters. CoRR abs\/1702.05373.","journal-title":"CoRR"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-44586-2_9"},{"key":"e_1_3_2_12_2","unstructured":"Dheeru Dua and Casey Graff. 2017. UCI Machine Learning Repository. Retreived from http:\/\/archive.ics.uci.edu\/ml."},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.2017.1389735"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1145\/1536414.1536467"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1561\/0400000042"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/TDSC.2019.2949041"},{"key":"e_1_3_2_17_2","article-title":"Federated learning for ranking browser history suggestions","volume":"1911","author":"Hartmann Florian","year":"2019","unstructured":"Florian Hartmann, Sunah Suh, Arkadiusz Komarzewski, Tim D. Smith, and Ilana Segall. 2019. Federated learning for ranking browser history suggestions. CoRR abs\/1911.11807.","journal-title":"CoRR"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/34.291440"},{"key":"e_1_3_2_19_2","first-page":"6343","volume-title":"Advances in Neural Information Processing Systems","author":"Jayaraman Bargav","year":"2018","unstructured":"Bargav Jayaraman, Lingxiao Wang, David Evans, and Quanquan Gu. 2018. Distributed learning without distress: Privacy-preserving empirical risk minimization. In Advances in Neural Information Processing Systems. Curran Associates Inc., 6343\u20136354."},{"issue":"2","key":"e_1_3_2_20_2","article-title":"Comprehensive analysis of privacy leakage in vertical federated learning during prediction","volume":"2022","author":"Jiang Xue","year":"2022","unstructured":"Xue Jiang, Xuebing Zhou, and Jens Grossklags. 2022. Comprehensive analysis of privacy leakage in vertical federated learning during prediction. Proc. Priv. Enhanc. Technol. 2022, 2 (2022). 263\u2013281.","journal-title":"Proc. Priv. Enhanc. Technol."},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.2478\/popets-2022-0024"},{"key":"e_1_3_2_22_2","article-title":"Model pruning enables efficient federated learning on edge devices","volume":"1909","author":"Jiang Yuang","year":"2019","unstructured":"Yuang Jiang, Shiqiang Wang, Bong-Jun Ko, Wei-Han Lee, and Leandros Tassiulas. 2019. Model pruning enables efficient federated learning on edge devices. CoRR abs\/1909.12326.","journal-title":"CoRR"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1561\/2200000083"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1137\/090756090"},{"key":"e_1_3_2_25_2","first-page":"202","volume-title":"Proceedings of the 2nd SIGKDD Conference on Knowledge Discovery and Data Mining","author":"Kohavi Ron","year":"1996","unstructured":"Ron Kohavi. 1996. Scaling up the accuracy of Naive-Bayes classifiers: A decision-tree hybrid. In Proceedings of the 2nd SIGKDD Conference on Knowledge Discovery and Data Mining. AAAI Press, 202\u2013207."},{"key":"e_1_3_2_26_2","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1989.1.4.541"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2020.2975749"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-32692-0_16"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2020.2986024"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-59410-7_33"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i08.7021"},{"key":"e_1_3_2_32_2","first-page":"1273","volume-title":"Proceedings of the 20th International Conference on Artificial Intelligence and Statistics","author":"McMahan Brendan","year":"2017","unstructured":"Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Ag\u00fcera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. In Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. PMLR, 1273\u20131282."},{"key":"e_1_3_2_33_2","volume-title":"Proceedings of the 6th International Conference on Learning Representations","author":"McMahan Brendan","year":"2018","unstructured":"Brendan McMahan, Daniel Ramage, Kunal Talwar, and Li Zhang. 2018. Learning differentially private recurrent language models. In Proceedings of the 6th International Conference on Learning Representations."},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/FOCS.2007.41"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2019.00065"},{"key":"e_1_3_2_36_2","article-title":"Collecting and analyzing data from smart device users with local differential privacy","volume":"1606","author":"Nguy\u00ean Th\u00f4ng T.","year":"2016","unstructured":"Th\u00f4ng T. Nguy\u00ean, Xiaokui Xiao, Yin Yang, Siu Cheung Hui, Hyejin Shin, and Junbum Shin. 2016. Collecting and analyzing data from smart device users with local differential privacy. CoRR abs\/1606.05053.","journal-title":"CoRR"},{"key":"e_1_3_2_37_2","volume-title":"Proceedings of the 5th International Conference on Learning Representations","author":"Papernot Nicolas","year":"2017","unstructured":"Nicolas Papernot, Mart\u00edn Abadi, \u00dalfar Erlingsson, Ian J. Goodfellow, and Kunal Talwar. 2017. Semi-supervised knowledge transfer for deep learning from private training data. In Proceedings of the 5th International Conference on Learning Representations."},{"key":"e_1_3_2_38_2","first-page":"185","volume-title":"Fast Training of Support Vector Machines Using Sequential Minimal Optimization","author":"Platt John C.","year":"1999","unstructured":"John C. Platt. 1999. Fast Training of Support Vector Machines Using Sequential Minimal Optimization. MIT Press, Cambridge, MA, 185\u2013208."},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1145\/2810103.2813687"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/MIS.2020.2993966"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1145\/3338501.3357370"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1145\/3378679.3394533"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE.2019.00063"},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM.2018.8486234"},{"key":"e_1_3_2_45_2","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1965.10480775"},{"key":"e_1_3_2_46_2","article-title":"Fashion-MNIST: A novel image dataset for benchmarking machine learning algorithms","volume":"1708","author":"Xiao Han","year":"2017","unstructured":"Han Xiao, Kashif Rasul, and Roland Vollgraf. 2017. Fashion-MNIST: A novel image dataset for benchmarking machine learning algorithms. CoRR abs\/1708.07747.","journal-title":"CoRR"},{"key":"e_1_3_2_47_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.101992"},{"key":"e_1_3_2_48_2","series-title":"Lecture Notes in Computer Science","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1007\/978-3-030-63076-8_16","volume-title":"Federated Learning\u2014Privacy and Incentive","author":"Yang Liu","year":"2020","unstructured":"Liu Yang, Ben Tan, Vincent W. Zheng, Kai Chen, and Qiang Yang. 2020. Federated recommendation systems. In Federated Learning\u2014Privacy and Incentive. Lecture Notes in Computer Science, Vol. 12500. Springer, 225\u2013239."},{"key":"e_1_3_2_49_2","article-title":"Applied federated learning: Improving Google keyboard query suggestions","volume":"1812","author":"Yang Timothy","year":"2018","unstructured":"Timothy Yang, Galen Andrew, Hubert Eichner, Haicheng Sun, Wei Li, Nicholas Kong, Daniel Ramage, and Fran\u00e7oise Beaufays. 2018. Applied federated learning: Improving Google keyboard query suggestions. CoRR abs\/1812.02903.","journal-title":"CoRR"},{"key":"e_1_3_2_50_2","first-page":"160","volume-title":"Proceedings of the 23rd Annual Symposium on Foundations of Computer Science","author":"Yao Andrew Chi-Chih","year":"1982","unstructured":"Andrew Chi-Chih Yao. 1982. Protocols for secure computations (extended abstract). In Proceedings of the 23rd Annual Symposium on Foundations of Computer Science. IEEE Computer Society, 160\u2013164."},{"issue":"4","key":"e_1_3_2_51_2","first-page":"25:1\u201325:41","article-title":"Privbayes: Private data release via Bayesian networks","volume":"42","author":"Zhang Jun","year":"2017","unstructured":"Jun Zhang, Graham Cormode, Cecilia M. Procopiuc, Divesh Srivastava, and Xiaokui Xiao. 2017. Privbayes: Private data release via Bayesian networks. ACM Trans. Datab. Syst. 42, 4 (2017), 25:1\u201325:41.","journal-title":"ACM Trans. Datab. Syst."},{"issue":"11","key":"e_1_3_2_52_2","first-page":"8836","article-title":"Local differential privacy-based federated learning for Internet of Things","volume":"8","author":"Zhao Yang","year":"2021","unstructured":"Yang Zhao, Jun Zhao, Mengmeng Yang, Teng Wang, Ning Wang, Lingjuan Lyu, Dusit Niyato, and Kwok-Yan Lam. 2021. Local differential privacy-based federated learning for Internet of Things. IEEE IoT J. 8, 11 (2021), 8836\u20138853.","journal-title":"IEEE IoT J."},{"key":"e_1_3_2_53_2","first-page":"14747","volume-title":"Proceedings of the 32nd Annual Conference on Neural Information Processing Systems","author":"Zhu Ligeng","year":"2019","unstructured":"Ligeng Zhu, Zhijian Liu, and Song Han. 2019. Deep leakage from gradients. In Proceedings of the 32nd Annual Conference on Neural Information Processing Systems. Curran Associates, Inc., 14747\u201314756."}],"container-title":["ACM Transactions on Intelligent Systems and Technology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3517820","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3517820","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T17:49:05Z","timestamp":1750182545000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3517820"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,11]]},"references-count":52,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,10,31]]}},"alternative-id":["10.1145\/3517820"],"URL":"https:\/\/doi.org\/10.1145\/3517820","relation":{},"ISSN":["2157-6904","2157-6912"],"issn-type":[{"value":"2157-6904","type":"print"},{"value":"2157-6912","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,11]]},"assertion":[{"value":"2021-03-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-02-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-06-11","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}