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 601–700 of 1,035
“Fairness for unobserved characteristics: insights from technological impacts on queer communities”.
Tomasev, N., McKee, K. R., Kay, J., & Mohamed, S. (2021) · Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3461702.3462540 · Cited by 70
“Promise Advances in privacy-preserving machine learning (Nasr, Shokri, and Houmansadr 2018; Bonawitz et al. 2017; Jayaraman and Evans 2019) present the possibility that the queer community might benefit from AI systems while minimising the risk of information leakage.”
BackgroundA robust analysis of adversarial attacks on federated learning environments
Nair, A. K., Raj, E. D., & Sahoo, J. (2023) · Computer Standards & Interfaces · doi:10.1016/j.csi.2023.103723 · Cited by 70
Fold-stratified cross-validation for unbiased and privacy-preserving federated learning
Bey, R., Goussault, R., Grolleau, F., Benchoufi, M., & Porcher, R. (2020) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocaa096 · Cited by 71
“This technique enables the training of AI models while keeping records in decentralized trusted data warehouses curated for instance by hospitals [19–25].”
MethodFederated Learning Under Importance Sampling
Rizk, E., Vlaski, S., & Sayed, A. H. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2022.3210365 · Cited by 70
FGFL: A blockchain-based fair incentive governor for Federated Learning
Gao, L., Li, L., Chen, Y., Xu, C., & Xu, M. (2022) · Journal of Parallel and Distributed Computing · doi:10.1016/j.jpdc.2022.01.019 · Cited by 71
Eiffel : Efficient and Fair Scheduling in Adaptive Federated Learning
Sultana, A., Haque, M. M., Chen, L., Xu, F., & Yuan, X. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2022.3187365 · Cited by 71
“To enable efficient federated learning in such a resourceconstrained environment, existing works have investigated a number of approaches ([3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], etc.”
BackgroundFirdaus, M., Larasati, H. T., & Rhee, K.-H. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101579 · Cited by 70
Zhao, Y., Chen, J., Zhang, J., Wu, D., Blumenstein, M., & Yu, S. (2022) · Concurrency and Computation: Practice and Experience · doi:10.1002/cpe.5906 · Cited by 59
“Furthermore,recentdefensemethodsmainlyfocusonencryptionandsecuremultipartycompu-tationtechniques, 12-14 Byzantine-tolerantlearning, 15,16 anomalydetection, 17 andclustering.”
LDP-Fed: Federated Learning with Local Differential PrivacyPreprint
Truex, S., Liu, L., Chow, K.-H., Gürsoy, M. E., & Wei, W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.03637 · Cited by 71
“Existing solutions to protect FL systems from such privacy attacks require trusted aggregators [17] or heavy cryptographic techniques [6, 23] which do not allow individual participants to define different local privacy guarantees, are insufficient for meaning-fully protecting each high dimensional…”
BackgroundEnhancing Privacy via Hierarchical Federated Learning
Wainakh, A., Guinea, A. S., Grube, T., & Mühlhäuser, M. (2020) · IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) · doi:10.1109/eurospw51379.2020.00053 · Cited by 69
“g, secure aggregation [2] or secure shuffling [1].”
BackgroundToward Secure and Privacy-Preserving Distributed Deep Learning in Fog-Cloud Computing
Li, Y., Li, H., Xu, G., Xiang, T., Huang, X., & Lu, R. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3012480 · Cited by 65
“and communication overhead, the SMC-based PPML [12] is compared with our SPDDL.”
BackgroundMethodResultInfluentialPrivacy-Preserving and Verifiable Federated Learning Framework for Edge Computing
Zhou, H., Yang, G., Huang, Y., Dai, H., & Xiang, Y. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3227435 · Cited by 69
“Moreover, [16] maintains the security even if a part of the clients drop out.”
BackgroundFedLGA: Toward System-Heterogeneity of Federated Learning via Local Gradient Approximation
Li, X., Qu, Z., Tang, B., & Lu, Z. (2023) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2023.3247365 · Cited by 71
“…model is trained on all remote devices in the FL network to solve an optimization problem without exchanging their private training data, which distinguishes the FL paradigm from traditional centralized machine learning, and thus the data privacy can be greatly protected [2], [4], [5], [6], [7].”
BackgroundFederated Learning in Practice: Reflections and Projections
Daly, K., Eichner, H., Kairouz, P., McMahan, H. B., Ramage, D., & Xu, Z. (2024) · IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA) · doi:10.1109/tps-isa62245.2024.00026 · Cited by 70
“Google’s cross-device federated learning system [10] features various synchronization points - in part to support the synchronous, round-based FedAvg learning algorithm [48], in part to aid data minimization by supporting the secure aggregation protocol [9].”
Self-citationToward Multiple Federated Learning Services Resource Sharing in Mobile Edge Networks
Nguyen, M. N. H., Tran, N. H. K., Tun, Y. K., Han, Z., & Hong, C. S. (2021) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2021.3085979 · Cited by 69
“In addition to Differential Privacy, Secure Aggregation [36] approaches focus on the aggregation operations using encryption techniques without revealing the contribution of each user.”
BackgroundTrustworthy AI in the Age of Pervasive Computing and Big Data
Kumar, A., Braud, T., Tarkoma, S., & Hui, P. (2020) · 2020 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops48775.2020.9156127 · Cited by 68
Multi-party Private Set Intersection in Vertical Federated Learning
Lu, L., & Ding, N. (2020) · IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · doi:10.1109/trustcom50675.2020.00098 · Cited by 68
PRECODE - A Generic Model Extension to Prevent Deep Gradient Leakage
Scheliga, D., Mäder, P., & Seeland, M. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00366 · Cited by 68
Federated learning attack surface: taxonomy, cyber defences, challenges, and future directions
Qammar, A., Ding, J., & Ning, H. (2021) · Artificial Intelligence Review · doi:10.1007/s10462-021-10098-w · Cited by 70
“Furthermore, implementation of bootstraps trust and assigning the test scores are helpful to cope with trust challenges Cao et al. (2020). Besides, the future may include a hierarchical root of trust structure that would include multiple subsets of samples at different trust levels.”
BackgroundFedHM: Efficient federated learning for heterogeneous models via low-rank factorization
Yao, D., Pan, W., Shi, Y., O'Neill, M. J., Dai, Y., Wan, Y., Zhao, P., Jin, H., & Sun, L. (2025) · Artificial Intelligence · doi:10.1016/j.artint.2025.104333 · Cited by 70
Privacy-Preserving Federated Learning Against Label-Flipping Attacks on Non-IID Data
Shen, X., Liu, Y., Li, F., & Li, C. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3288886 · Cited by 68
“Specifically, we use the standard hybrid argument [40], [41] to prove that during the execution of the algorithm, the view of any server (PS Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
BackgroundMethodInfluentialFedCRMW: Federated model ownership verification with compression-resistant model watermarking
Nie, H., & Lu, S. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.123776 · Cited by 68
Federated learning: A cutting-edge survey of the latest advancements and applications
Akhtarshenas, A., Vahedifar, M. A., Ayoobi, N., Maham, B., Alizadeh, T., Ebrahimi, S., & López-Pérez, D. (2024) · Computer Communications · doi:10.1016/j.comcom.2024.107964 · Cited by 68
Location Privacy in Usage-Based Automotive Insurance: Attacks and Countermeasures
Zhou, L., Du, S., Zhu, H., Chen, C., Ota, K., & Dong, M. (2018) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2018.2848227 · Cited by 68
“Then, we refer to the proof that was proposed in [20] and [21] to provide the security proof of our scheme.”
BackgroundMethodMLGuard: Mitigating Poisoning Attacks in Privacy Preserving Distributed Collaborative Learning
Khazbak, Y., Tan, T., & Cao, G. (2020) · 2020 29th International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn49398.2020.9209670 · Cited by 39
“al [9] proposed a collaborative learning solution which leverages secure multiparty computation to aggregate all parameter updates obtained from users’ devices without revealing users’ updates in the clear.”
BackgroundTowards Crowdsourced Training of Large Neural Networks using Decentralized Mixture-of-Experts
Ryabinin, M., & Gusev, A. (2020) · arXiv · doi:10.48550/arxiv.2002.04013
“Secure aggregation procedures (Bonawitz et al., 2017) require multiple workers to communicate and scale quadratically with their number.”
BackgroundProof of Learning (PoLe): Empowering neural network training with consensus building on blockchains
Liu, Y., Lan, Y., Li, B., Miao, C., & Tian, Z. (2021) · Computer Networks · doi:10.1016/j.comnet.2021.108594 · Cited by 67
Ge, L., Li, H., Wang, X., & Wang, Z. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126897 · Cited by 68
FLShield: A Validation Based Federated Learning Framework to Defend Against Poisoning Attacks
Kabir, E., Song, Z., Rashid, M. R. U., & Mehnaz, S. (2024) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp54263.2024.00141 · Cited by 67
“…Attacks Distributed Backdoor Attack (DBA) [13] Edge-case Backdoor Attack (ECBA) [52] Semantic Backdoor Attack (SBA) [2] Privacy Inference Attack Gradient Inversion Attack (GIA) [17] FLShield -aware Attacks FLShield -aware adaptive attack (FA-Adp) FLShield -aware advanced attack (FA-Adv) i ∈ [ c…”
BackgroundBlockchain-Based Privacy-Enhancing Federated Learning in Smart Healthcare: A Survey
Limbepe, Z. N., Gai, K., & Yu, J. (2025) · Blockchains · doi:10.3390/blockchains3010001 · Cited by 68
FedGT: Identification of Malicious Clients in Federated Learning With Secure Aggregation
Xhemrishi, M., Östman, J., Wachter-Zeh, A., & Amat, A. G. I. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3539964 · Cited by 68
“, 2017) and FL with full secure aggregation (Bonawitz et al., 2017) can be seen as the two extreme cases of FedGT, corresponding to n (non-overlapping) groups and a single group with n clients, respectively.”
Towards Private, Robust, and Verifiable Crowdsensing Systems via Public Blockchains
Cai, C., Zheng, Y., Du, Y., Qin, Z., & Wang, C. (2019) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2019.2941481 · Cited by 67
Privacy for Free: Communication-Efficient Learning with Differential Privacy Using SketchesPreprint
Li, T., Liu, Z., Sekar, V., & Smith, V. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.00972 · Cited by 66
“Previous privacy-preserving methods typically add noise to guarantee differential privacy [35], or use cryptographic protocols such as secure multi-party computation (SMC) [6].”
BackgroundBlockFLow: An Accountable and Privacy-Preserving Solution for Federated LearningPreprint
Mugunthan, V., Rahman, R., & Kagal, L. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.03856 · Cited by 69
“Differential privacy [15], the process of adding random noise to the model parameters [6, 8, 31], prevents these attacks.”
BackgroundCommunication-Efficient Large-Scale Distributed Deep Learning: A Comprehensive SurveyPreprint
Liang, F., Zhang, Z., Lu, H., Leung, V. C. M., Guo, Y., & Hu, X. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2404.06114 · Cited by 69
“The scheme adopts the Secure Aggregation protocol [140], performing model aggregation in a logical incorruptible third party.”
MethodA survey of local differential privacy for securing internet of vehicles
Zhao, P., Zhang, G., Wan, S., Liu, G., & Umer, T. (2020) · The Journal of Supercomputing · doi:10.1007/s11227-019-03104-0 · Cited by 66
“Moreover, National Institute of Standards and Technology of United States has developed the “Safety Framework for Vehicle Network Attack Protection” in 2014 [10, 12].”
BackgroundRobustFL: Robust Federated Learning Against Poisoning Attacks in Industrial IoT Systems
Zhang, J., Ge, C., Hu, F., & Chen, B. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3132954 · Cited by 66
Scalable and Low-Latency Federated Learning With Cooperative Mobile Edge Networking
Zhang, Z., Gao, Z., Guo, Y., & Gong, Y. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3216837 · Cited by 68
“Furthermore, CE-FedAvg is compatible with existing privacy-preserving techniques in FL such as secure aggregation [26], [27], differential privacy [28], [29], [30], and shuffling [31] since only the sum rather than individual values is needed for the intra-cluster and inter-cluster model…”
MethodOn the (In)security of Peer-to-Peer Decentralized Machine Learning
Pasquini, D., Raynal, M., & Troncoso, C. (2023) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp46215.2023.10179291 · Cited by 66
“Assuming fault resilient SA [4] (which is necessary under real-world deployments), this strategy would work also in a complete topology, where N(Aa)=N(Ab).”
BackgroundFedFM: Anchor-Based Feature Matching for Data Heterogeneity in Federated Learning
Ye, R., Ni, Z., Xu, C., Wang, J., Chen, S., & Eldar, Y. C. (2023) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2023.3314277 · Cited by 67
“Meanwhile, Secure Aggregation [60] is often used in practice to ensure the safety of model parameters, which can also be adopted to further secure the anchors communication.”
BackgroundIOFL: Intelligent-Optimization-Based Federated Learning for Non-IID Data
Li, X., Zhao, H., & Deng, W. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3354942 · Cited by 66
“Bonawitz et al. [34] proposed a secure aggregation strategy based on secret sharing, which can prevent servers from obtaining client model parameter information while completing parameter aggregation.”
BackgroundChen, J., Wang, E. Z., Srivastava, G., Alghamdi, T. A., Khan, F., Kumari, S., & Xiong, H. (2024) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2024.100593 · Cited by 68
Differentially Private Federated Stochastic Primal-Dual Learning for Internet of Vehicles
Li, Y., Wang, S., & Chang, T. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3534982 · Cited by 74
“…have been used in FL to provide the privacy guarantee, such as secure multiparty computation (SMC) [21], homomorphic encryption [22], and secret sharing [23], these methods require complicated encryption protocols and computation, imposing prohibitive overheads on local clients and the PS.”
Transparent Contribution Evaluation for Secure Federated Learning on Blockchain
Ma, S., Cao, Y., & Xiong, L. (2021) · IEEE International Conference on Data Engineering Workshops (ICDEW) · doi:10.1109/icdew53142.2021.00023 · Cited by 67
“2) To address the above privacy issue, we can adopt the well-studied privacy-preserving federated learning, such as secure aggregation (which masks updates) [7].”
Distributed Sensing Using Smart End-User Devices: Pathway to Federated Learning for Autonomous IoT
Imteaj, A., & Amini, M. H. (2019) · 2019 International Conference on Computational Science and Computational Intelligence (CSCI) · doi:10.1109/csci49370.2019.00218 · Cited by 65
“The fourth one is, Privacy concerns [23], which is very evident to arise in federated learning as the local devices are sharing their model updates periodically.”
BackgroundShuffleFL: gradient-preserving federated learning using trusted execution environment
Zhang, Y., Wang, Z.-W., Cao, J., Hou, R., & Meng, D. (2021) · ACM International Conference on Computing Frontiers (CF) · doi:10.1145/3457388.3458665 · Cited by 65
“Although secure aggregation ensures no individual participant’s gradients can be inspected, it is confronted with poor robustness issues [3].”
BackgroundVPFL: A verifiable privacy-preserving federated learning scheme for edge computing systems
Zhang, J., Liu, Y., Wu, D., Lou, S., Chen, B., & Yu, S. (2022) · Digital Communications and Networks · doi:10.1016/j.dcan.2022.05.010 · Cited by 65
Tiansuan Constellation: An Open Research Platform
Wang, S., Li, Q., Xu, M., Ma, X., Zhou, A., & Sun, Q. (2021) · IEEE International Conference on Edge Computing and Communications (EDGE) · doi:10.1109/edge53862.2021.00022 · Cited by 65
Source Inference Attacks: Beyond Membership Inference Attacks in Federated Learning
Hu, H., Zhang, X., Salcic, Z. A., Sun, L., Choo, K. R., & Dobbie, G. (2023) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2023.3321565 · Cited by 66
“These privacy-preserving methods [22], [60], [61], [91], [95], [97] have been evaluated and demonstrated to be effective for mitigating privacy attacks in FL.”
BackgroundMethodSecure and efficient federated learning via novel multi-party computation and compressed sensing
Chen, L., Xiao, D., Yu, Z., & Zhang, M. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120481 · Cited by 66
Does Differential Privacy Really Protect Federated Learning From Gradient Leakage Attacks?
Hu, J., Du, J., Wang, Z., Pang, X., Zhou, Y., Sun, P., & Ren, K. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3417930 · Cited by 66
“Specifically, FL uses homomorphic encryption [32], [33] or secret sharing [34], [35] in MPC technologies to achieve secure aggregation, which can protect individual gradients from server access.”
BackgroundPrivacy-Preserving Outsourced Speech Recognition for Smart IoT Devices
Ma, Z., Liu, Y., Liu, X., Ma, J., & Li, F. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2917933 · Cited by 59
“As for the security model, we adopt the standard semihonest security model [36], which is also called passive or honest-but-curious model.”
MethodExploring Collection of Sign Language Datasets: Privacy, Participation, and Model Performance
Bragg, D., Koller, O., Caselli, N., & Thies, W. (2020) · Proceedings of the 22nd International ACM SIGACCESS Conference on Computers and Accessibility · doi:10.1145/3373625.3417024 · Cited by 44
“Techniques exist to aggregate high-dimensional data from multiple sources without revealing individual contributions [9], and to learn many types of models including decision trees [3, 78], linear regression [35] Naive Bayes classifers [115], and k-means clustering [64].”
BackgroundFedEval: A Holistic Evaluation Framework for Federated LearningPreprint
Chai, D., Wang, L., Liu, Y., Zhang, J., Chen, K., & Yang, Q. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.09655 · Cited by 63
A Survey on Vertical Federated Learning: From a Layered PerspectivePreprint
Liu, Y., Chai, D., Zhang, J., Jin, Y., Wang, L., Liu, H., Han, T., Xu, Q., & Chen, K. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2304.01829 · Cited by 65
“The secure aggregation method [8] adopted by [163, 166] could also solve the dropout problems by utilizing SS.”
MethodInfluentialA survey for federated learning evaluations: Goals and measures
Chai, D., Wang, L., Yang, L., Zhang, J., Chen, K., & Yang, Q. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3382002 · Cited by 64
“Another example that demonstrates the necessity of considering communication and computation simultaneously in the efficiency evaluation is when comparing the efficiency of two different privacy protection techniques: SS [49], [64], [76], [92] and HE [46], [92].”
BackgroundMethodFederated Learning with Personalized Local Differential Privacy
Yang, G., Wang, S., & Wang, H. (2021) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs52626.2021.9449232 · Cited by 64
“Multiple solutions have been proposed to solve this problem, such as FL based on homomorphic encryption(HE)[6] or secure multi-party computation(SMC) [7], they are both lossless but high-cost.”
BackgroundPoster: FLATEE: Federated Learning Across Trusted Execution Environments
Mondal, A., More, Y., Rooparaghunath, R. H., & Gupta, D. (2021) · IEEE European Symposium on Security and Privacy · doi:10.1109/eurosp51992.2021.00054 · Cited by 64
“Popular FL frameworks such as [2,14] often incur large amounts of training time and communication latencies (due to the computationally intensive cryptographic operations involved).”
When Machine Learning Meets Spectrum Sharing Security: Methodologies and Challenges
Wang, Q., Sun, H., Hu, R. Q., & Bhuyan, A. (2022) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2022.3146364 · Cited by 65
Li, Y., & Mao, H. (2022) · Scientific Programming · doi:10.1155/2022/3317876 · Cited by 64
Fed-LTD: Towards Cross-Platform Ride Hailing via Federated Learning to Dispatch
Wang, Y., Tong, Y., Zhou, Z., Ren, Z., Xu, Y., Wu, G., & Lv, W. (2022) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3534678.3539047 · Cited by 64
“To obfuscate which grid has been updated, we use random masking [2] to preserve the privacy of value updates.”
MethodHow Much Privacy Does Federated Learning with Secure Aggregation Guarantee?
Elkordy, A. R., Zhang, J., Ezzeldin, Y. H., Psounis, K., & Avestimehr, S. (2023) · Proceedings on Privacy Enhancing Technologies · doi:10.56553/popets-2023-0030 · Cited by 64
Mistry, D., Mridha, M. F., Safran, M., Alfarhood, S., Saha, A. K., & Che, D. (2023) · IEEE Access · doi:10.1109/access.2023.3299331 · Cited by 64
Efficient Data Collaboration Using Multi-Party Privacy Preserving Machine Learning Framework
Salam, A., Abrar, M. Y. B. T., Ullah, F., Khan, I. A., Amin, F., & Choi, G. S. (2023) · IEEE Access · doi:10.1109/access.2023.3339750 · Cited by 64
“Communication between the central server and the devices occurs via encrypted channels, often leveraging secure and efficient communication protocols [10].”
BackgroundRFed: Robustness-Enhanced Privacy-Preserving Federated Learning Against Poisoning Attack
Miao, Y., Yan, X., Li, X., Xu, S., Liu, X., Li, H., & Deng, R. H. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3402113 · Cited by 65
“…mechanism Cheon-Kim-Kim-Song (CKKS), rather than Differential Privacy (DP) [16], [17], Homomorphic Encryption (HE) [18], [19], [20], [21], [22] and Secure Multi-party Computation (SMC) [23], [24], due to low model accuracy, high computation overhead and communication overhead, respectively.”
BackgroundPublicly verifiable privacy-preserving aggregation and its application in IoT
Li, T., Gao, C., Jiang, L., Pedrycz, W., & Shen, J. (2018) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2018.09.018 · Cited by 64
On lightweight privacy-preserving collaborative learning for internet-of-things objects
Jiang, L., Tan, R., Lou, X., & Lin, G. (2019) · Proceedings of the International Conference on Internet of Things Design and Implementation · doi:10.1145/3302505.3310070 · Cited by 57
“In [36] and [9], homomorphic encryption and secure aggregation have been applied to enhance the privacy preservation of the approach in [40] and the federated learning in [31], respectively.”
BackgroundMethodFoundations of Cryptoeconomic Systems
Voshmgir, S., & Zargham, M. (2019) · ePubWU Institutional Repository (Wirtschaftsuniversität Wien) · doi:10.57938/0d57c21f-6e9d-4e94-89ff-ec2d4d64583e · Cited by 64
Partitioned Variational Inference: A unified framework encompassing federated and continual learningPreprint
Bui, T. D., Nguyen, C. V., Swaroop, S., & Turner, R. E. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.11206 · Cited by 64
“It can also be further improved by employing encrypted aggregation steps [Bonawitz et al., 2017] or differentially-private mechanisms [Dwork and Roth, 2014].”
BackgroundInfluentialFedJAX: Federated learning simulation with JAXPreprint
Ro, J. H., Suresh, A. T., & Wu, K. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.02117 · Cited by 64
Zhou, J., Chen, S., Wu, Y., Li, H., Zhang, B., Zhou, L., Hu, Y., Xiang, Z., Li, Z., Chen, N., et al. (2024) · Science Advances · doi:10.1126/sciadv.adh8601 · Cited by 63
Zhao, J., Bagchi, S., Avestimehr, S., Chan, K., Chaterji, S., Dimitriadis, D., Li, J., Li, N., Nourian, A., & Roth, H. (2025) · ACM Computing Surveys · doi:10.1145/3724113 · Cited by 63
“[23,26,73,121,150] Datareconstruction(suitablemainlyforcross-device) Requiresadditionalclient-to-clientcommunicationforencryptionmasks.”
InfluentialPoulain, R., Bin Tarek, M. F., & Beheshti, R. (2023) · 2023 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3593013.3594102 · Cited by 63
“…the model’s parameters as the weighted sum of the locally trained parameters: Following this method, each client solely has to communicate its 𝐹 score alongside its respective number of training instances which can be done without privacy concerns through methods such as Secure Aggregation [43].”
A privacy preserving federated learning scheme using homomorphic encryption and secret sharing
Shi, Z., Yang, Z., Hassan, A., Li, F., & Ding, X. (2022) · Telecommunication Systems · doi:10.1007/s11235-022-00982-3 · Cited by 63
Chen, J., Zhao, Y., Li, Q. L., Feng, X., & Xu, K. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3297369 · Cited by 65
“1) Datasets: We choose four network intrusion attack datasets to evaluate our defense, including KDDCUP’99 [38] (abbreviated as KDD99), Mirai botnet [12], CIC-IDS2017 [30] (abbreviated as CIC2017), and UNSW-NB15 [40] (abbreviated as UNSW).”
BackgroundInfluentialYe, Z., Zhang, X., Chen, X., Xiong, H., & Yu, D. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2023.3312511 · Cited by 63
“With the secure aggregation protocol [47], namely one-time pad masking, the parameters within each cluster can be aggregated as: in which θ k denotes the averaged model parameters of the k -th cluster, and s i,j , s j,i are some random vectors to protect the parameters.”
MethodByzantine-Robust Federated Learning with Optimal Statistical Rates
Zhu, B., Wang, L., Pang, Q., Wang, S., Jiao, J., Song, D., & Jordan, M. I. (2023) · Rare & Special e-Zone (The Hong Kong University of Science and Technology) · Cited by 63
Applications of machine learning in cryptography: a survey
Alani, M. M. (2019) · International Conference on Cryptography, Security and Privacy (CSP) · doi:10.1145/3309074.3309092 · Cited by 63
Privacy-preserving Data Mining in Industry
Kenthapadi, K., Mironov, I., & Thakurta, A. (2019) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3289600.3291384 · Cited by 63
“In the last part of the tutorial, we will discuss emerging topics pertaining to privacy-preserving data mining and machine learning such as distributed private machine learning [5, 21–23], encode-shuffle-analyze architecture [4], and privacy amplification [13].”
BackgroundA survey on privacy inference attacks and defenses in cloud-based Deep Neural Network
Zhang, X., Chen, C., Xie, Y., Chen, X., Zhang, J., & Xiang, Y. (2022) · Computer Standards & Interfaces · doi:10.1016/j.csi.2022.103672 · Cited by 64
A Framework for Edge-Assisted Healthcare Data Analytics using Federated Learning
Hakak, S., Ray, S., Khan, W. Z., & Scheme, E. (2020) · 2020 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata50022.2020.9377873 · Cited by 62
“For secure local model and global update aggregation, homomorphic encryption [6] could be used, as it allows calculations to be computed over the encrypted data without the need for decryption.”
MethodScalable and Differentially Private Distributed Aggregation in the Shuffled Model
Ghazi, B., Pagh, R., & Velingker, A. (2019) · arXiv · doi:10.48550/arxiv.1906.08320 · Cited by 62
“Current practical secure aggregation protocols such as that of Bonawitz et al. [6] Single-user change”
BackgroundMethodFedCDR: Federated Cross-Domain Recommendation for Privacy-Preserving Rating Prediction
Wu, M., Li, L., Chang, T., Rigall, E., Xiao-dong, W., & Cheng-Zhong, X. (2022) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3511808.3557320 · Cited by 62
“In that case, the security of transmittable parameters (including the item embedding parameters and transfer parameters) can be strengthened through secure multi-party computation (MPC), homomorphic encryption [2] and secure aggregation algorithm [4].”
BackgroundPrivacy-preserving federated k-means for proactive caching in next generation cellular networks
Liu, Y., Ma, Z., Yan, Z., Wang, Z., Liu, X., & Ma, J. (2020) · Information Sciences · doi:10.1016/j.ins.2020.02.042 · Cited by 55
A Critical Evaluation of Privacy and Security Threats in Federated Learning
Asad, M., Moustafa, A., & Yu, C. (2020) · Sensors · doi:10.3390/s20247182 · Cited by 51
“FedCS [19] HFL NN centralized general BlockFL [20] HFL LM centralized general Tree-based FL [21] HFL DT distributed DP Adversarial Lens [22] HFL NN centralized general Secure Aggregation [23] HFL NN centralized CT FedXGB [24] HFL DT centralized CT SecureBoost [25] VFL DT centralized CT q-FedAvg [26] HFL NN, LM central…”
BackgroundFog-Embedded Deep Learning for the Internet of Things
Lyu, L., Bezdek, J. C., He, X., & Jin, J. (2019) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2019.2912465 · Cited by 61
“and unbalanced data, including FedAvg and FedSGD developed by Google [2], [12].”
MethodBiscotti: A Ledger for Private and Secure Peer-to-Peer Machine Learning
Shayan, M., Fung, C., Yoon, C. J. M., & Beschastnikh, I. (2018) · arXiv · doi:10.48550/arxiv.1811.09904 · Cited by 61
“Additional security work in federated learning has required that public key infrastructure exists which validates the identity of users [12], but prior sybil work has shown that relying on public key infrastructure for user validation is insufficient [52, 53].”
BackgroundInfluentialLi, J., Cui, T., Yang, K., Yuan, R., He, L., & Li, M. (2021) · Sustainability · doi:10.3390/su132313050 · Cited by 61
Zhang, K., Wang, Y., Wang, H., Huang, L., Yang, C., Chen, X., & Sun, L. (2022) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2022.findings-emnlp.43 · Cited by 61
Efficient Participant Contribution Evaluation for Horizontal and Vertical Federated Learning
Wang, J., Zhang, L., Li, A., You, X., & Cheng, H. (2022) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde53745.2022.00073 · Cited by 61
Clustered Federated Multitask Learning on Non-IID Data With Enhanced Privacy
Shu, J., Yang, T., Liao, X., Chen, F., Xiao, Y., Yang, K., & Jia, X. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3228893 · Cited by 62
“The first practical secure aggregation in FL was proposed based on threshold secret sharing, and it is also robust to a fraction of user dropouts [27].”
MethodFalcon: A Privacy-Preserving and Interpretable Vertical Federated Learning System
Wu, Y., Xing, N., Chen, G., Dinh, T. T. A., Luo, Z., Ooi, B. C., Xiao, X., & Zhang, M. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3603581.3603588 · Cited by 61
“FL can be classified into three settings depending on how the data is partitioned, namely horizontal FL [8, 50], vertical FL [27, 34, 46, 73], and hybrid [75].”
BackgroundFederated Noisy Client Learning
Tam, K., Li, L., Han, B., Xu, C., & Fu, H. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3336050 · Cited by 61
“[4] designed a communication-efficient, failure-robust protocol for secure aggregation of high-dimensional data.”
BackgroundFederated Object Detection: Optimizing Object Detection Model with Federated Learning
Yu, P., & Liu, Y. (2019) · Proceedings of the 3rd International Conference on Vision, Image and Signal Processing · doi:10.1145/3387168.3387181 · Cited by 44
“Bonawitz et al. [10] presented a communication-efficient and robust protocol, and make it a reality to aggregate high-dimensional data from multiple mobile devices securely.”
BackgroundFederated learning and privacy
Bonawitz, K., Kairouz, P., McMahan, H. B., & Ramage, D. (2022) · Communications of the ACM · doi:10.1145/3500240 · Cited by 62
“5 Such algorithms are both: ˲ Communication efficient -”
MethodSelf-citationFederated Survival Analysis with Discrete-Time Cox ModelsPreprint
Andreux, M., Manoel, A., Menuet, R., Saillard, C., & Simpson, C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.08997 · Cited by 62
“In order to prevent data leakages, techniques such as differential privacy [11, 2] and secure aggregation [3] can be used.”
MethodPractical and Robust Federated Learning With Highly Scalable Regression Training
Han, S., Ding, H., Zhao, S., Ren, S., Wang, Z., Lin, J., & Zhou, S. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3271859 · Cited by 60
“How to design a robust NFRT scheme is challenging [9], [16], [17], [18].”
BackgroundMethodInfluentialSecure and Privacy-Preserving Federated Learning via Co-Utility
Domingo‐Ferrer, J., Blanco-Justicia, A., Manjón, J., & Sánchez, D. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3102155 · Cited by 60
Enhanced Security and Privacy via Fragmented Federated Learning
Jebreel, N., Domingo‐Ferrer, J., Blanco-Justicia, A., & Sánchez, D. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3212627 · Cited by 60
“Bonawitz et al. [32] and So et al. [33] used secret sharing to hide individual updates.”
BackgroundMethodFLASH: Heterogeneity-Aware Federated Learning at Scale
Yang, C., Xu, M., Wang, Q., Chen, Z., Huang, K., Ma, Y., Bian, K., Huang, G., Liu, Y., Jin, X., et al. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3214234 · Cited by 62
“Besides FedAvg , various aggregation algorithms have been proposed to improve effi-ciency [17], [18], [48], ensure fairness [13], preserve privacy [36], [48], etc.”
BackgroundMethod
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.