Practical Secure Aggregation for Privacy-Preserving Machine Learning
The DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.
Abstract
We design a novel, communication-efficient, failure-robust protocol for secure aggregation of high-dimensional data. Our protocol allows a server to compute the sum of large, user-held data vectors from mobile devices in a secure manner (i.e. without learning each user's individual contribution), and can be used, for example, in a federated learning setting, to aggregate user-provided model updates for a deep neural network. We prove the security of our protocol in the honest-but-curious and active adversary settings, and show that security is maintained even if an arbitrarily chosen subset of users drop out at any time. We evaluate the efficiency of our protocol and show, by complexity analysis and a concrete implementation, that its runtime and communication overhead remain low even on large data sets and client pools. For 16-bit input values, our protocol offers $1.73 x communication expansion for 210 users and 220-dimensional vectors, and 1.98 x expansion for 214 users and 224-dimensional vectors over sending data in the clear.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Although SMC guarantees that none of the parties shares anything with each other or with any third party, it can not prevent an adversary from learning some individual information”
From Federated Learning for Healthcare Informatics (2021) · cites this paper as [8]
In the citing paper
“By combining local inference with trusted aggregation, TDAI can enhance both privacy and robustness of anomaly detection pipelines [239].”
From Anomaly detection and root-cause identification in microservices: a survey (2026) · cites this paper as [239]
In the citing paper
“FL is still in its infancy and is still vulnerable to various challenges [148,149].”
From Smart Wearables for the Detection of Cardiovascular Diseases: A Systematic Literature Review (2023) · cites this paper as [148]
In the citing paper
“Therefore, many FL frameworks use secure aggregation [81], [82], [83], [84] algorithms in the aggregation process, at which point most of the poisoning attack defense methods mentioned above will be invalid.”
From Poisoning attacks in federated learning: A survey (2023) · cites this paper as [81]
In the citing paper
“Bonawitz et al. [62] introduce a secure aggregation protocol for protecting individual model updates.”
From Handling privacy-sensitive medical data with federated learning: Challenges and future directions (2023) · cites this paper as [62]
6 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
New directions in cryptography
Diffie, W., & Hellman, M. (1976) · IEEE Transactions on Information Theory · doi:10.1109/tit.1976.1055638 · Cited by 14,713
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215 · cited 1×
“maintains a private database of her text messages securely on her own mobile device, and a shared global model is trained under the coordination of a central server based upon highly processed, minimally scoped, ephemeral updates from users [20], [10].”
BackgroundSelf-citationShamir, A. (1979) · Communications of the ACM · doi:10.1145/359168.359176 · Cited by 13,756 · cited 1×
“We rely on Shamir’s t -out-of- n Secret Sharing [38], which allows a user to split a secret s into n shares, such that any t shares can be used to reconstruct s , but any set of at most t − 1 shares gives no information about s .”
Method(2016) · Cited by 11,679
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578 · cited 3×
“This is particular advantageous in the local privacy setting [13], which offers provable guarantees for the protection of individual training examples [14], [15] even when the data aggregator is not assumed to be trusted [16], [17].”
BackgroundMethodSelf-citationThe Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557
Dwork, C. (2006) · Lecture Notes in Computer Science · doi:10.1007/11787006_1 · Cited by 5,291 · cited 1×
“In such cases, secure aggregation composes well with differential privacy [12].”
BackgroundMembership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
Federated Learning: Strategies for Improving Communication Efficiency
Konečný, J., McMahan, H. B., Yu, F. X., Richtárik, P., Suresh, A. T., & Bacon, D. (2016) · arXiv · doi:10.48550/arxiv.1610.05492 · Cited by 3,766
Self-citation
Research works printed but not linked: 3
- Stephen Advokat. 1987. Publication Of Bork's Video Rentals Raises Privacy Issue. Chicago Tribune (1987).
- Michael Barbaro, Tom Zeller, and Saul Hansell. 2006. A face is exposed for AOL searcher no. 4417749. New York Times 9, 2008 (2006).
- Elaine Shi, HTH Chan, Eleanor Rieffel, Richard Chow, and Dawn Song. 2011. Privacy-preserving aggregation of time-series data. In Annual Network & Distributed System Security Symposium (NDSS). Internet Society.
Cited by
Items 701–800 of 1,035
SVFL: Efficient Secure Aggregation and Verification for Cross-Silo Federated Learning
Luo, F., Al-Kuwari, S. M., & Ding, Y. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3219485 · Cited by 60
“Based on the framework proposed by Bonawitz et al. [8], Bell et al. [7] and Choi et al. [10] proposed secure aggregation protocols with polylogarithmic communication and computation overheads.”
BackgroundMethodInfluentialPrivacy-Preserving and Reliable Decentralized Federated Learning
Gao, Y., Zhang, L., Wang, L., Choo, K. R., & Zhang, R. (2023) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2023.3250705 · Cited by 61
“Similar to the scheme in [25], it is the synchronous FL scheme with the highest comprehensive efficiency that ensures full data privacy.”
BackgroundMethodResultInfluentialMaize yield prediction using federated random forest
Zhang, Q., Zhao, X., Han, Y., Yang, F., Pan, S., Liu, Z., Wang, K., & Zhao, C. (2023) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2023.107930 · Cited by 60
FedBEVT: Federated Learning Bird's Eye View Perception Transformer in Road Traffic Systems
Song, R., Xu, R., Festag, A., Ma, J., & Knoll, A. (2023) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2023.3310674 · Cited by 60
A Survey on the use of Federated Learning in Privacy-Preserving Recommender Systems
Chronis, C., Varlamis, I., Himeur, Y., Sayed, A. N., Al-Hasan, T. M., Nhlabatsi, A., Bensaali, F., & Dimitrakopoulos, G. J. (2024) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2024.3396344 · Cited by 62
“On the other hand, in a cross-device setting, stability is often compromised due to factors such as mobile devices with poor connections [40] or limited computational resources.”
BackgroundPrivacy-preserving blockchain-based federated learning for brain tumor segmentation
Kumar, R., Bernard, C. M., Ullah, A., Khan, R. U., Kumar, J., Kulevome, D. K. B., Rao, Y., & Zeng, S. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108646 · Cited by 61
SERVERLESS INTELLIGENCE: SECURING J2EE-BASED FEDERATED LEARNING PIPELINES ON AWS
Oleti, C. S. (2022) · INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING & TECHNOLOGY · doi:10.34218/ijcet_13_03_017 · Cited by 60
PADL: Privacy-Aware and Asynchronous Deep Learning for IoT Applications
Liu, X., Li, H., Xu, G., Liu, S., Liu, Z., & Lu, R. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2981379 · Cited by 61
“Particularly, the mini-batch stochastic gradient descent method [37] is also used to optimize the model as many state-of-the-art works [38], [39].”
MethodA Verifiable Privacy-Preserving Machine Learning Prediction Scheme for Edge-Enhanced HCPSs
Li, X., He, J., Vijayakumar, P., Zhang, X., & Chang, V. I. C. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3110808 · Cited by 60
Secure and robust machine learning for healthcare: A survey
Qayyum, A., Qadir, J., Bilal, M., & Al-Fuqaha, A. (2021) · IEEE Reviews in Biomedical Engineering · doi:10.1109/rbme.2020.3013489 · Cited by 59
“In a similar study [142], a protocol is developed using the “secret sharing” strategy for aggregating model updates received from multiple input parties, the updates are used for training of the ML model.”
MethodSecure and Robust Machine Learning for Healthcare: A Survey
Qayyum, A., Qadir, J., Bilal, M., & Al-Fuqaha, A. (2020) · arXiv · doi:10.48550/arxiv.2001.08103 · Cited by 59
“In a similar study [142], a protocol is developed using the “secret sharing” strategy for aggregating model updates received from multiple input parties, the updates are used for training of the ML model.”
MethodFederated Multi-view Learning for Private Medical Data Integration and Analysis
Che, S., Kong, Z., Peng, H. M., Sun, L., Leow, A., Chen, Y., & He, L. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3501816 · Cited by 59
“designed a secure aggregation scheme that allows a server to securely compute users’ data from mobile devices [5].”
BackgroundContribution‐based Federated Learning client selection
Lin, W., Xu, Y., Liu, B., Li, D., Huang, T., & Shi, F. (2022) · International Journal of Intelligent Systems · doi:10.1002/int.22879 · Cited by 60
“To protect individual model updates, Bonawitz et al. 13 introduce multiparty secure computing protocols.”
MethodSecure Shapley Value for Cross-Silo Federated Learning
Zheng, S., Cao, Y., & Yoshikawa, M. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3587136.3587141 · Cited by 61
MPCFL: Towards Multi-party Computation for Secure Federated Learning Aggregation
Kaminaga, H., Awaysheh, F. M., Alawadi, S., & Kamm, L. (2023) · IEEE/ACM International Conference on Utility and Cloud Computing (UCC) · doi:10.1145/3603166.3632144 · Cited by 59
In the Pursuit of Privacy: The Promises and Predicaments of Federated Learning in Healthcare
Topaloglu, M. Y., Morrell, E. M., Rajendran, S., & Topaloglu, U. (2021) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2021.746497 · Cited by 50
VCD-FL: Verifiable, Collusion-Resistant, and Dynamic Federated Learning
Gao, S., Luo, J., Zhu, J., Dong, X., & Shi, W. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3271268 · Cited by 58
“Existing works [18], [20], [22] have proposed to subtract off the masks before gradient aggregation with the double-masking protocol.”
InfluentialFlGan: GAN-Based Unbiased Federated Learning Under Non-IID Settings
Ma, Z., Liu, Y., Miao, Y., Xu, G., Liu, X., Ma, J., & Deng, R. H. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2023.3309858 · Cited by 58
“In a non-IID scenario, the global distribution is composed of distinct local data distributions p i ∈ [1 ,n ] .”
BackgroundChougule, S., Chaudhari, B. S., Ghorpade, S. N., & Zennaro, M. (2024) · World Electric Vehicle Journal · doi:10.3390/wevj15020039 · Cited by 58
Privacy-Preserving distributed deep learning based on secret sharing
Duan, J., Zhou, J., & Li, Y. (2020) · Information Sciences · doi:10.1016/j.ins.2020.03.074 · Cited by 58
Pokhrel, S. R. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3060185 · Cited by 59
Practical Lossless Federated Singular Vector Decomposition over Billion-Scale Data
Chai, D., Wang, L., Zhang, J., Liu, Y., Cai, S., Chen, K., & Yang, Q. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3539402 · Cited by 58
“1) According to the prior work [3], the CSP only learns the aggregated results X′, no information is leaked during the secure aggregation; 2) In Theorem 2, we show that there is an infinite number of raw data that could be masked into the same matrix.”
BackgroundBlockchain-Federated-Learning and Deep Learning Models for COVID-19 Detection Using CT Imaging
Kumar, R., Khan, A. A., Kumar, J., Zakria, Golilarz, N. A., Zhang, S., Ting, Y., Zheng, C., & Wang, W. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3076767 · Cited by 57
A Joint Energy and Latency Framework for Transfer Learning Over 5G Industrial Edge Networks
Yang, B. F., Fagbohungbe, O. I., Cao, X., Yuen, C., Qian, L., Niyato, D. T., & Zhang, Y. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3075444 · Cited by 57
Security of Federated Learning: Attacks, Defensive Mechanisms, and Challenges
Benmalek, M., Benrekia, M. A., & Challal, Y. (2022) · Revue d intelligence artificielle · doi:10.18280/ria.360106 · Cited by 57
“Although SMC-based methods [60-64] provide a secure aggregation of the protected clients’ updates, they induce significant extra communication overhead among clients which may be unaffordable for some devices and networks.”
Nguyen, M.-D., Lee, S., Pham, Q., Hoang, D. T., Nguyen, D. N., & Hwang, W. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3190510 · Cited by 57
Dai, S., & Meng, F. (2022) · Applied Intelligence · doi:10.1007/s10489-022-04065-3 · Cited by 59
“[36] designed a secure aggregation scheme to further enhance the privacy of aggregated intermediate updates based on [34].”
BackgroundEGIA: An External Gradient Inversion Attack in Federated Learning
Liang, H., Li, Y., Zhang, C., Liu, X., & Zhu, L. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3302161 · Cited by 57
Liu, J., Zhang, J., Jan, M. A., Sun, R. X., Liu, L., Verma, S., & Chatterjee, P. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3304361 · Cited by 57
“Bonawitz et al. [23] constructed a secure aggregation protocol via double-masking and secret sharing, and it can sustain robustness in the case of network delay and user dropout.”
BackgroundPersonalized and privacy-preserving federated heterogeneous medical image analysis with PPPML-HMI
Zhou, J., Zhou, L., Wang, D., Xu, X., Li, H., Chu, Y., Han, W., & Gao, X. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107861 · Cited by 57
AutoReP: Automatic ReLU Replacement for Fast Private Network Inference
Peng, H., Huang, S., Zhou, T., Luo, Y., Wang, C., Wang, Z., Zhao, J., Xi, X., Li, A., Geng, T., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00478 · Cited by 59
“The existing highly-secure PI solutions usually use cryptographic primitives include multi-party computation (MPC) [16, 4, 15] and homomorphic encryption (HE) [23, 5, 11, 26].”
BackgroundNeural Data Server: A Large-Scale Search Engine for Transfer Learning Data
Yan, X., Acuna, D., & Fidler, S. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00395 · Cited by 56
Practical Private Aggregation in Federated Learning Against Inference Attack
Zhao, P., Cao, Z., Jiang, J., & Gao, F. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3201231 · Cited by 57
“Other kinds of works [16], [17], [18], [19] apply SMC to FL settings to enable worker devices to encrypt and/or secret-share their local models to the honest-but-curious FL server.”
BackgroundMethodRobust Learning from Untrusted SourcesPreprint
Konstantinov, N., & Lampert, C. H. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.10310 · Cited by 57
“In cases when the gradients may reveal sensitive information about the data, secure aggregation (Bonawitz et al., 2017) or other privacy-preserving distributed learning methods (Shokri & Shmatikov, 2015) can be used on top to ensure privacy.”
MethodFederated Learning for Object Detection in Autonomous Vehicles
Jallepalli, D., Ravikumar, N. C., Badarinath, P. V., Uchil, S., & Suresh, M. A. (2021) · 2021 IEEE Seventh International Conference on Big Data Computing Service and Applications (BigDataService) · doi:10.1109/bigdataservice52369.2021.00018 · Cited by 56
Federated learning to comply with data protection regulations
Chalamala, S. R., Kummari, N. K., Singh, A. K., Saibewar, A., & Chalavadi, K. M. (2022) · CSI Transactions on ICT · doi:10.1007/s40012-022-00351-0 · Cited by 56
“To prevent the server from estimating individual data samples based on the summary update it has received, a Secure aggregation protocol [38] is developed by Google.”
MethodA federated learning framework for pneumonia image detection using distributed data
Kareem, A., Liu, H. M., & Velisavljević, V. (2023) · Healthcare Analytics · doi:10.1016/j.health.2023.100204 · Cited by 56
Khalil, S. S., Tawfik, N. S., & Spruit, M. R. (2024) · Applied Intelligence · doi:10.1007/s10489-023-05095-1 · Cited by 57
“To ensure privacy, a secure aggregation protocol was developed, allowing the server to decrypt the average update only if a predetermined number of users have participated and sent their results [13].”
Intelligent Edge Computing and Machine Learning: A Survey of Optimization and Applications
Cajas, S. A., Samanta, J., Suárez‐Cetrulo, A. L., & Carbajo, R. S. (2025) · Future Internet · doi:10.3390/fi17090417 · Cited by 57
“The future of intelligent machine learning at the edge is shaped by cloud offloading techniques, distributed learning approaches, context-awareness, soft computing model compression techniques, specialized hardware, and security and privacy via design principles [199–202].”
Dual Defense: Enhancing Privacy and Mitigating Poisoning Attacks in Federated Learning
Xu, R., Gao, S., Li, C., Joshi, J., & Li, J. (2024) · neural information processing systems · doi:10.52202/079017-2253 · Cited by 58
Towards Differentially Private Text Representations
Lyu, L., Li, Y., He, X., & Xiao, T. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401260 · Cited by 55
“The other cryptograph-based methods can be resource-hungry or overly complex for users [2].”
BackgroundContamination Attacks and Mitigation in Multi-Party Machine Learning
Hayes, J., & Ohrimenko, O. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.02402 · Cited by 35
“Secure MPC can be enabled with cryptographic techniques [6, 11, 18, 29, 32], and systems based on trusted processors such as Intel SGX [3, 5, 33].”
Hybrid Differentially Private Federated Learning on Vertically Partitioned DataPreprint
Wang, C., Liang, J., Huang, M., Bai, B., Bai, K., & Li, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.02763 · Cited by 56
“The numerous participants of HFL are mostly personal smart phones or edge devices with limited computation power and network bandwidth, thus the goal of gradient protection in HFL setting is mainly achieved by differential privacy (DP)[14, 1, 2, 26, 33], secret sharing[5], and so on.”
BackgroundMethodInfluentialFedGraphNN: A Federated Learning Benchmark System for Graph Neural Networks
He, C., Balasubramanian, K., Ceyani, E., Yang, C., Xie, H., Sun, L., He, L., Yang, L., Yu, P. S., Rong, Y., et al. (2021) · Cited by 56
FedCDR: Privacy-preserving federated cross-domain recommendation
Yan, D., Zhao, Y., Yang, Z., Jin, Y., & Zhang, Y. (2022) · Digital Communications and Networks · doi:10.1016/j.dcan.2022.04.034 · Cited by 55
Towards energy-aware federated learning on battery-powered clients
Arouj, A., & Abdelmoniem, A. M. (2022) · doi:10.1145/3556557.3557952 · Cited by 57
“, shorter training rounds) [31, 43] while other mechanisms have been proposed to guarantee privacy and robustness [15, 19].”
BackgroundBoenisch, F., Dziedzic, A., Schuster, R., Shamsabadi, A. S., Shumailov, I., & Papernot, N. (2023) · IEEE European Symposium on Security and Privacy · doi:10.1109/eurosp57164.2023.00023 · Cited by 56
“Since SA-protocols provide their guarantees under the assumption that a certain fraction of users is honest [6], [10], it follows naturally that in the presence of the sybil devices, no guarantees can be provided to the target user.”
BackgroundCommunications Security in Industry X: A Survey
Ahmad, I., Rodriguez, F., Kumar, T., Suomalainen, J., Jagatheesaperumal, S. K., Walter, S., Asghar, M. Z., Li, G., Papakonstantinou, N., Ylianttila, M., et al. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3356076 · Cited by 57
“The challenges of preserving data privacy in machine learning to maintain company information or factory floor plans can be addressed with privacy-preserving federated learning approaches, such as discussed in [147].”
BackgroundDecentralised, collaborative, and privacy-preserving machine learning for multi-hospital data
Fang, C., Dziedzic, A., Zhang, L., Oliva, L., Verma, A. A., Razak, F. A., Papernot, N., & Wang, B. (2024) · EBioMedicine · doi:10.1016/j.ebiom.2024.105006 · Cited by 55
“For the patient survival/mortality prediction using GEMINI dataset, which is a binary classi fi cation task, we evaluate the Area under the receiver operating characteristic curve Classifying cell types.”
BackgroundMethodA review on client-server attacks and defenses in federated learning
Sharma, A., & Marchang, N. (2024) · Computers & Security · doi:10.1016/j.cose.2024.103801 · Cited by 56
Federated Learning for Human Activity Recognition: Overview, Advances, and Challenges
Aouedi, O., Sacco, A., Khan, L. U., Nguyen, D. C., & Guizani, M. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3484228 · Cited by 57
Dubey, P., & Kumar, M. (2024) · Computer Science Review · doi:10.1016/j.cosrev.2024.100697 · Cited by 56
Data Leakage in Federated Averaging
Dimitrov, D. I., Balunović, M., Konstantinov, N., & Vechev, M. (2022) · Transactions on machine learning research · doi:10.3929/ethz-b-000593553 · Cited by 56
FedMEC: Improving Efficiency of Differentially Private Federated Learning via Mobile Edge Computing
Zhang, J., Zhao, Y., Wang, J., & Chen, B. (2020) · Mobile Networks and Applications · doi:10.1007/s11036-020-01586-4 · Cited by 55
“Furthermore, Bonawitz et al. proposed a sum aggregation protocol for high-dimensional data using the secret-sharing method [28].”
MethodApplying Federated Learning in Software-Defined Networks: A Survey
Ma, X., Liao, L., Li, Z., Lai, R. X., & Zhang, M. (2022) · Symmetry · doi:10.3390/sym14020195 · Cited by 55
“[98] designed a secure communication protocol that could specify a constant number of iterations and tolerate a number of clients exiting the FL.”
BackgroundFederated matrix factorization with privacy guarantee
Li, Z., Ding, B., Zhang, C., Li, N., & Zhou, J. (2021) · Proceedings of the VLDB Endowment · doi:10.14778/3503585.3503598 · Cited by 55
Gradient-Leaks: Understanding and Controlling Deanonymization in Federated LearningPreprint
Orekondy, T., Oh, S. J., Zhang, Y., Schiele, B., & Fritz, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.05838 · Cited by 65
“While initial research [7, 22, 48, 65] has been conducted on keeping the data private, risks about linkability have been largely overlooked.”
BackgroundSAFELearning: Secure Aggregation in Federated Learning With Backdoor Detectability
Zhang, Z., Li, J., Yu, S., & Makaya, C. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3280032 · Cited by 54
“Intuitively, secure aggregation with the tree structure provides similar level of protection to model privacy of users as in [10], and the user dropouts can be handled similarly as well.”
BackgroundMethodInfluentialAveraging Is Probably Not the Optimum Way of Aggregating Parameters in Federated Learning
Xiao, P., Cheng, S., Stankovic, V., & Vukobratovic, D. (2020) · Entropy · doi:10.3390/e22030314 · Cited by 54
“Some research has focused on further enhancing the privacy using 37 various cryptographic protocols[19][20], or applying differential privacy[21] to confuse attackers.”
BackgroundThe Applications of Blockchain in Artificial Intelligence
Wang, R., Luo, M., Wen, Y., Wang, L., Choo, K. R., & He, D. (2021) · Security and Communication Networks · doi:10.1155/2021/6126247 · Cited by 54
Collaborative Machine Learning: Schemes, Robustness, and Privacy
Wang, J., Pal, A., Yang, Q., Kant, K., Zhu, K., & Guo, S. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3169347 · Cited by 54
“…Referring to the HTTPs communication algorithm to make the updated parameters invisible to the third party , and add the special noise to it [141] Making a comprehensive analytic of the Model Confidence and the Gradient Training two kinds of model, one is trained with the dataset having the…”
BackgroundMethodInfluentialFederated Boosted Decision Trees with Differential Privacy
Maddock, S., Cormode, G., Wang, T., Maple, C. R., & Jha, S. (2022) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3548606.3560687 · Cited by 54
“A powerful tool is secure aggregation, which allows the computation of a sumwithout revealing any intermediate values [7, 10].”
MethodBlockchain-Based Decentralized Model Aggregation for Cross-Silo Federated Learning in Industry 4.0
Ranathunga, T., McGibney, A., Rea, S., & Bharti, S. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3218704 · Cited by 54
“Although, a compromised centralized server can be dealt with the presence of a secure aggregator [5] or using a secure local model sharing technique such as homomorphic encryption (HE), clients can still be compromised or can behave selfishly and share low-quality local model updates with the server.”
BackgroundResultInfluentialFederated Learning Over Coupled Graphs
Lei, R., Wang, P., Zhao, J., Lan, L., Tao, J., Deng, C., Feng, J., Wang, X., & Guan, X. (2023) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2023.3240527 · Cited by 54
“In addition, previous works, such as [18], [19] providing secure aggregation methods can also be used to preserve privacy in the training steps of our FedCog.”
Shao, J., Li, Z., Sun, W., Zhou, T., Sun, Y., Liu, L., Lin, Z., Mao, Y., & Zhang, J. (2026) · ACM Computing Surveys · doi:10.1145/3844940 · Cited by 56
“However, directly adding noise or quantizing local predictions may introduce a substantial tradeoff between privacy budget and model performance.”
Zhang, J., Zhu, C., Ge, C., Ma, C., Zhao, Y., Sun, X., & Chen, B. (2024) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2024.3354049 · Cited by 54
“Moreover, these defense strategies are incompatible withsecureaggregation[15],[16],[17],whicharewidelyusedin FL. Thesecondtypeofdefensestrategyisbackdoormitigation which aims to purify the backdoored model through different perspectives.”
BackgroundEfficiently Achieving Privacy Preservation and Poisoning Attack Resistance in Federated Learning
Li, X.-Y., Yang, X. S., Zhou, Z., & Lu, R. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3378006 · Cited by 54
“However, the large-scale collection of sensitive data poses a significant risk of privacy leakage [1].”
BackgroundMethodA Verifiable Privacy-Preserving Federated Learning Framework Against Collusion Attacks
Chen, Y., He, S., Wang, B., Feng, Z., Zhu, G., & Tian, Z. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3516119 · Cited by 54
“Bonawitz et al. [23] introduced a secure data aggregation approach utilizing secret sharing techniques, which permited any participant to revoke their involvement.”
Kalakoti, R., Nõmm, S., & Bahşi, H. (2025) · Computer Networks · doi:10.1016/j.comnet.2025.111479 · Cited by 55
Toward Verifiable and Privacy Preserving Machine Learning Prediction
Niu, C., Wu, F., Tang, S., Ma, S., & Chen, G. (2020) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2020.3035591 · Cited by 54
“VerifyNet not only guarantees the privacy of the model updates with the celebrated secure aggregation protocol in [57] but also ensures the correctness of the aggregated result with homomorphic hash function.”
BackgroundA Verifiable Federated Learning Scheme Based on Secure Multi-party Computation
Mou, W.-H., Fu, C., Yan, L., & Hu, C. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86130-8_16 · Cited by 54
Artificial Identification: A Novel Privacy Framework for Federated Learning Based on Blockchain
Ouyang, L., Wang, F., Tian, Y., Jia, X., Qi, H., & Wang, G. (2023) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2022.3226861 · Cited by 54
“Many works have been proposed to reduce the communication costs [1], [13], [14] and strengthen the aggregation security [15], [16], [17], [18], [19] in the collaborative group.”
BackgroundAdversarial training in communication constrained federated learningPreprint
Shah, D. P., Dube, P., Chakraborty, S., & Verma, A. K. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.01319 · Cited by 54
“While federation allows the clients to retain control and privacy of their raw data [2, 26], the final model continues to remain vulnerable to carefully crafted adversarial examples [25, 3, 8] designed to elicit misclassification at inference time.”
Label Leakage and Protection from Forward Embedding in Vertical Federated LearningPreprint
Sun, J., Yang, X., Yao, Y., & Wang, C. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2203.01451 · Cited by 54
“There are three main categories of information protection techniques in vFL: 1) cryptographic methods such as secure multi-party computation [2]; 2) system-based methods including trusted execution environments [15]; and 3) perturbation methods that add noise to the communicated messages [1, 14, 7, 5, 23].”
BackgroundNon-interactive verifiable privacy-preserving federated learning
Xu, Y., Peng, C., Tan, W., Tian, Y., Ma, M., & Niu, K. (2021) · Future Generation Computer Systems · doi:10.1016/j.future.2021.10.017 · Cited by 53
FLeet: Online Federated Learning via Staleness Awareness and Performance Prediction
Damaskinos, G., Guerraoui, R., Kermarrec, A.-M., Nitu, V., Patra, R., & Taïani, F. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3527621 · Cited by 53
“Differential privacy [26] is a popular technique for privacy-preserving FL with formal guarantees [9].”
Efficient and Secure Federated Learning With Verifiable Weighted Average Aggregation
Yang, Z., Zhou, M. L., Yu, H., Sinnott, R., & Liu, H. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3206243 · Cited by 53
“To solve the gradient leakage problem that may be caused by communication security threats [23], [24], [25], researchers adopt many methods to protect the participant’s gradient and local model [26], [27], [28], [29], [30], [31], [32].”
Prompt-enhanced Federated Content Representation Learning for Cross-domain Recommendation
Guo, L., Lu, Z., Yu, J., Nguyen, Q. V. H., & Yin, H. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645337 · Cited by 53
“Recently, several studies have been focused on conducting privacy preserving in CDR tasks [1, 2, 10, 25].”
BackgroundTraining Machine Learning Models at the Edge: A Survey
Khouas, A. R., Bouadjenek, M. R., Hacid, H., & Aryal, S. (2026) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3848509 · Cited by 56
Security, privacy, and robustness for trustworthy AI systems: A review
Saeed, M. M., & Alsharidah, M. (2024) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2024.109643 · Cited by 56
Wen, H., Liu, X., Lei, B., Yang, M., Cheng, X., & Chen, Z. (2024) · Applied Energy · doi:10.1016/j.apenergy.2024.124789 · Cited by 54
Towards compute-efficient Byzantine-robust federated learning with fully homomorphic encryption
Jiang, S., Yang, H., Xie, Q., Ma, C., Wang, S., Liu, Z., Xiang, T., & Xing, G. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01107-6 · Cited by 53
“Unlike differential privacy that can directly protect data privacy 15,16 , cryptographic frameworks such as multi-party computation 17 , partially homomorphic encryption (HE) 18 and fully homomorphic encryption (FHE) 19 have been developed to enable secure computations and prevent unauthorized…”
InfluentialAn Efficient Blockchain-Based Privacy-Preserving Collaborative Filtering Architecture
Casino, F., & Patsakis, C. (2019) · IEEE Transactions on Engineering Management · doi:10.1109/tem.2019.2944279 · Cited by 53
“Users may use the protocol described in [61] so that they connect with a server by using a secure channel and compute the aggregate value of their evaluations.”
MethodTimeCrypt: Encrypted Data Stream Processing at Scale with Cryptographic Access Control
Burkhalter, L., Hithnawi, A., Viand, A., Shafagh, H., & Ratnasamy, S. (2018) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000402391 · Cited by 53
“This can be accomplished by a secure aggregation protocol [6, 19] between the involved users and the analyst.”
BackgroundMethodAsynchronous Federated Learning with Differential Privacy for Edge IntelligencePreprint
Li, Y., Yang, S., Ren, X., & Zhao, C. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.07902 · Cited by 54
“To enhance the privacy guarantee of FL, many different techniques have been leveraged to prevent the indirect leakage, such as secure multiparty computation [8], homomorphic encryption [28], secret sharing [33], and differential privacy [30].”
MethodExploring privacy measurement in federated learning
Jagarlamudi, G. K., Yazdinejad, A., Parizi, R. M., & Pouriyeh, S. (2024) · The Journal of Supercomputing · doi:10.1007/s11227-023-05846-4 · Cited by 52
“Their descriptions are presented below: Secure Aggregation: Secure Aggregation is a privacy-preserving method [24–26] that encrypts and aggregates participant model updates.”
Privacy-aware and Resource-saving Collaborative Learning for Healthcare in Cloud Computing
Hao, M., Li, H., Xu, G., Liu, Z., & Chen, Z. (2020) · ICC 2020 - 2020 IEEE International Conference on Communications (ICC) · doi:10.1109/icc40277.2020.9148979 · Cited by 52
“6, in terms of the time of encryption, decryption and aggregation, we compare our PRCL with traditional privacy preserving federated learning (called PPFL), such as [10] [7].”
BackgroundMethodInfluentialBoosting Privately: Privacy-Preserving Federated Extreme Boosting for Mobile Crowdsensing
Liu, Y., Ma, Z., Liu, X., Ma, S., Nepal, S., & Deng, R. (2019) · arXiv · doi:10.48550/arxiv.1907.10218 · Cited by 52
“B. Keith et al. [17] designed a universal and practical model aggregation scheme for mobile devices with secret sharing technology.”
BackgroundA Survey and Guideline on Privacy Enhancing Technologies for Collaborative Machine Learning
Soykan, E. Ü., Karaçay, L., Karakoç, F., & Tomur, E. (2022) · IEEE Access · doi:10.1109/access.2022.3204037 · Cited by 52
“For the federated learningmethod, [31] proposed a solution 1402 to prevent the server from learning the weight updates by 1403 obscuring the aggregation from the server.”
BackgroundMethodSecure Summation: Capacity Region, Groupwise Key, and Feasibility
Zhao, Y., & Sun, H. (2023) · IEEE Transactions on Information Theory · doi:10.1109/tit.2023.3342571 · Cited by 52
“This work is particularly motivated by the secure aggregation problem [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], which arises recently in federated learning and the core is to securely compute the sum of the inputs available at a number of users without revealing any additional Yizhou Zhao…”
BackgroundIs Aggregation the Only Choice? Federated Learning via Layer-wise Model Recombination
Hu, M., Yue, Z., Xie, X., Chen, C., Huang, Y., Wei, X. H., Lian, X., Liu, Y., & Chen, M. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671722 · Cited by 53
Performance Analysis of Distributed and Federated Learning Models on Private Data
Chandiramani, K., Garg, D., & Maheswari, N. (2019) · Procedia Computer Science · doi:10.1016/j.procs.2020.01.039 · Cited by 52
Using Blockchain Technologies to Improve Security in Federated Learning Systems
Short, A. R., Leligou, H. C., Papoutsidakis, M., & Theocharis, E. (2020) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac48688.2020.00-96 · Cited by 53
FeARH: Federated machine learning with anonymous random hybridization on electronic medical records
Cui, J., Zhu, H., Deng, H., Chen, Z., & Liu, D. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103735 · Cited by 43
“[4] Bonawitz, Keith, Vladimir Ivanov, Ben Kreuter, Antonio Marcedone, H.”
BackgroundG-VCFL: Grouped Verifiable Chained Privacy-Preserving Federated Learning
Zhang, Z., Wu, L., He, D., Wang, Q., Wu, D., Shi, X., & Ma, C. (2022) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2022.3196404 · Cited by 51
PQLM - Multilingual Decentralized Portable Quantum Language Model
Li, S. S., Zhang, X., Zhou, S., Shu, H., Liang, R., Liu, H., & Garcia, L. P. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095215 · Cited by 51
“Even then, it is possible to leak sensitive information to a third party [21].”
BackgroundFL4IoT: IoT Device Fingerprinting and Identification Using Federated Learning
Wang, H., Eklund, D., Oprea, A., & Raza, S. (2023) · ACM Transactions on Internet of Things · doi:10.1145/3603257 · Cited by 51
“[3] designed a secure protocol, and McMahan et al.”
MethodWang, R., Yuan, X., Yang, Z., Wan, Y., Luo, M., & Wu, D. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.102029 · Cited by 51
Federated learning for anomaly detection on Internet of Medical Things: A survey
Pinto, R. P., Silva, B. M. C., & Inácio, P. R. M. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101677 · Cited by 52
Metadata
- Authors
- Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., Ramage, D., Segal, A., & Seth, K.
- Affiliations
- Google (United States) (US)
- Cornell Tech (US)
- Venue
- Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security
- Publisher
- ACM
- Published
- 2017-10-30
- Volume
- 2017
- Pages
- 1175-1191
- ISSN
- 1543-7221 (ISSN-L)
- OpenAlex
- W2767079719
- Cited by
- 3,956
- References
- 55
- Citations by year
- 2026: 701 · 2025: 712 · 2024: 643 · 2023: 564 · 2022: 400 · 2021: 436 · 2020: 299 · 2019: 152 · 2018: 42 · 2017: 7
- FWCI
- 62.37
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM SIGSAC Conference on Computer and Communications Security (CCS) (publishedVersion)
- IACR Cryptology ePrint Archive (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Information Systems › Cloud Data Security Solutions
- Keywords
- secure aggregation, privacy-preserving machine learning, federated learning, high-dimensional data, communication efficiency, user dropout, honest-but-curious adversary, active adversaries, secure multi-party computation
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.