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 201–300 of 1,035
Privacy-preserving federated learning for residential short-term load forecasting
Fernández, J. D., Menci, S. P., Lee, C. M., Rieger, A., & Fridgen, G. (2022) · Applied Energy · doi:10.1016/j.apenergy.2022.119915 · Cited by 196
“One such protocol is SecAgg [74].”
MethodA Survey on Federated Recommendation Systems
Sun, Z., Xu, Y., Liu, Y., He, W., Kong, L., Wu, F., Jiang, Y., & Cui, L. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3354924 · Cited by 194
“by using a secure aggregation protocol [79].”
Guo, P., Wang, P., Zhou, J., Jiang, S., & Patel, V. M. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00245 · Cited by 193
“Federated learning is a decentralized learning framework which allows multiple institutions to collaboratively learn a shared machine learning model without sharing their local training data [2, 25, 27].”
BackgroundFlexible Clustered Federated Learning for Client-Level Data Distribution Shift
Duan, M., Liu, D., Ji, X., Wu, Y., Liang, L., Chen, X., Tan, Y., & Ren, A. (2021) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3134263 · Cited by 128
“previous FedAvg-based frameworks [2], [15], [26], [35], [36], the optimization gradients in the server will not be updated or broadcast to all clients or groups in each round, we only maintain these gradients for the cold start of newcomers.”
BackgroundFedMultimodal: A Benchmark for Multimodal Federated Learning
Feng, T., Bose, D., Zhang, T., Hebbar, R., Ramakrishna, A., Gupta, R., Zhang, M., Avestimehr, S., & Narayanan, S. S. (2023) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3580305.3599825 · Cited by 193
“Apart from identifying privacy risks associated with multimodal FL, it is also crucial to investigate privacyenhancing techniques, such as differential privacy [17, 24, 72] and secure aggregation [6] as promising areas of research within the scope of FedMultimodal to mitigate privacy attacks.”
BackgroundLearning from History for Byzantine Robust OptimizationPreprint
Karimireddy, S. P., He, L., & Jaggi, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.10333 · Cited by 196
“, 2016), and is compatible with secure aggregation for federated learning (Bonawitz et al., 2017).”
BackgroundInfluentialDistributed Learning without Distress: Privacy-Preserving Empirical Risk Minimization
Jayaraman, B., Wang, L., Evans, D., & Gu, Q. (2018) · neural information processing systems · Cited by 192
Fast-adapting and privacy-preserving federated recommender system
Wang, Q., Yin, H., Chen, T., Yu, J., Zhou, A., & Zhang, X. (2022) · The VLDB Journal · doi:10.1007/s00778-021-00700-6 · Cited by 145
“Some simple functions have been implemented based on SMC, however, arbitrarily complex function computations such as deep learning are still infeasible due to computational cost [45].”
MethodInfluentialXie, Q., Jiang, S., Jiang, L., Huang, Y., Zhao, Z., Khan, S., Dai, W., Liu, Z., & Wu, K. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3382875 · Cited by 191
“MPC enables the involved parties to jointly compute a mutual function to aggregate the local models, i.e., garbled circuit (GC) [23], Masking [19] and secret sharing (SS) [24], [25].”
BackgroundDP-ADMM: ADMM-Based Distributed Learning With Differential Privacy
Huang, Z., Hu, R., Guo, Y., Chan-Tin, E., & Gong, Y. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2931068 · Cited by 190
“Some of them employ cryptography-based methods in the protocol to hide the private information [33]–[35].”
MethodLocal and Central Differential Privacy for Robustness and Privacy in Federated Learning
Naseri, M., Hayes, J., & De Cristofaro, E. (2022) · Proceedings 2022 Network and Distributed System Security Symposium · doi:10.14722/ndss.2022.23054 · Cited by 188
“Homomorphic Encryption (HE) can be used to encrypt participants’ parameters in such a way that the server can only decrypt the aggregates [9, 19, 39].”
BackgroundByrd, D., & Polychroniadou, A. (2020) · Proceedings of the First ACM International Conference on AI in Finance · doi:10.1145/3383455.3422562 · Cited by 188
“In particular, our secure weighted average protocol running across 𝑛 clients is based on the protocol of Bonawitz et al [3].”
MethodInfluentialA Privacy-Preserving Federated Learning for Multiparty Data Sharing in Social IoTs
Yin, L., Feng, J., Xun, H., Sun, Z., & Cheng, X. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3074185 · Cited by 188
Federated learning in smart cities: Privacy and security survey
Al-Huthaifi, R., Li, T., Huang, W. X., Gu, J., & Li, C. (2023) · Information Sciences · doi:10.1016/j.ins.2023.03.033 · Cited by 191
ELSA: Secure Aggregation for Federated Learning with Malicious Actors
Rathee, M., Shen, C., Wagh, S., & Popa, R. A. (2023) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp46215.2023.10179468 · Cited by 190
“Over the years, a rich line of work [19, 21, 22, 26, 35, 37, 40, 57, 62, 66, 70, 82, 84, 103, 105, 108, 115] has identified two security properties as desirable: • Privacy of individual gradients.”
BackgroundMethodInfluentialTowards Efficient and Privacy-Preserving Federated Deep Learning
Hao, M., Li, H., Xu, G., Liu, S., & Yang, H. (2019) · ICC 2019 - 2019 IEEE International Conference on Communications (ICC) · doi:10.1109/icc.2019.8761267 · Cited by 187
“Bonawitz et al. [17] proposed the first privacy-preserving federated deep learning approach which is robust to users dropping out by utilizing secure multiparty computation(MPC).”
BackgroundEncrypted Control for Networked Systems: An Illustrative Introduction and Current Challenges
Darup, M. S., Alexandru, A. B., Quevedo, D. E., & Pappas, G. J. (2021) · IEEE Control Systems · doi:10.1109/mcs.2021.3062956 · Cited by 188
“Furthermore, secure federated learning by Google [1], [11] exploits private aggregation to train a model from a multitude of different agents while reducing the amount of information collected.”
BackgroundEluding Secure Aggregation in Federated Learning via Model Inconsistency
Pasquini, D., Francati, D., & Ateniese, G. (2022) · Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3548606.3560557 · Cited by 104
“To increase the security of FL and mitigate attacks such as gradient inversion (see Section 2), Bonawitz et al. [10] propose a communication-efficient, dropout resilient SA for FL. 5 When applied to FL, the protocol guarantees a secure aggregation of users’ model updates – sensitive values that…”
BackgroundInfluentialAl-Quraan, M. M., Mohjazi, L., Bariah, L., Centeno, A., Zoha, A., Arshad, K., Assaleh, K. T., Muhaidat, S., Debbah, M. A., & Imran, M. A. (2023) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2023.3251404 · Cited by 188
“1) Secure multi-party computation (SMC): A cryptographic protocol that aims to conceal personal information and guarantee zero-knowledge between multiple involved parties [197], [198].”
BackgroundDifferential privacy-enabled federated learning for sensitive health data
Choudhury, O., Gkoulalas-Divanis, A., Salonidis, T., Sylla, I., Park, Y., Hsu, G., & Das, A. (2019) · arXiv · doi:10.48550/arxiv.1910.02578 · Cited by 184
“However, in certain cases, FL may be vulnerable to inference attacks [17, 18, 19].”
BackgroundFederated Evaluation of On-device Personalization
Wang, K., Mathews, R., Kiddon, C., Eichner, H., Beaufays, F., & Ramage, D. (2019) · arXiv · doi:10.48550/arxiv.1910.10252 · Cited by 184
“FL can be further combined with other privacy-preserving techniques like secure multi-party computation (Bonawitz et al., 2017) and differential privacy (McMahan et al., 2018; Agarwal et al., 2018; Abadi et al., 2016b).”
BackgroundSelf-citationPrivacy preserving vertical federated learning for tree-based models
Wu, Y., Cai, S., Xiao, X., Chen, G., & Ooi, B. C. (2020) · Proceedings of the VLDB Endowment · doi:10.14778/3407790.3407811 · Cited by 184
“Existing work on FL has mainly focused on the horizontal setting [7,8,47,48,50,53,54,60,75], which assumes that each client’s data have the same schema, but no tuple is shared by multiple clients.”
BackgroundHelen: Maliciously Secure Coopetitive Learning for Linear Models
Zheng, W., Popa, R. A., Gonzalez, J. E., & Stoica, I. (2019) · 2019 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2019.00045 · Cited by 134
“Coopetitive analytics systems [6, 55, 12, 21, 10] allow multiple parties to run SQL queries over all parties’ data.”
BackgroundFLOWER: A FRIENDLY FEDERATED LEARNING FRAMEWORKPreprint
Beutel, D. J., Topal, T., Mathur, A., Qiu, X., Fernández-Marqués, J., Gao, Y., Sani, L., Li, K. H., Parcollet, T., de Gusmão, P. P. B., et al. (2022) · HAL · Cited by 183
cpSGD: Communication-efficient and differentially-private distributed SGD
Agarwal, N., Suresh, A. T., Yu, F., Kumar, S., & McMahan, H. B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.10559 · Cited by 180
“, [7] adds log2(k · n) for k levels of quantization), our proposed communication benefits still hold.”
BackgroundInfluentialSelf-citationPrivacy Threat and Defense for Federated Learning With Non-i.i.d. Data in AIoT
Xiong, Z., Cai, Z., Takabi, D., & Li, W. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2021.3073925 · Cited by 180
“To protect data privacy in FL, secure multi-party computation (SMC) [5] and differential privacy (DP) [6] are commonly used solutions.”
BackgroundMethodAdaCliP: Adaptive Clipping for Private SGD
Pichapati, V., Suresh, A. T., Yu, F. X., Reddi, S. J., & Kumar, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.07643 · Cited by 60
“Even though noisy SGD usually provides global differential privacy, recent works have shown that they can be combined with cryptographic homomorphic encryption techniques to provide stronger privacy guarantees [30, 31].”
BackgroundSEAR: Secure and Efficient Aggregation for Byzantine-Robust Federated Learning
Zhao, L., Jiang, J., Feng, B. Y., Wang, Q., Shen, C. Q., & Li, Q. L. (2021) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3093711 · Cited by 180
“Like prior federated learning work [4], we consider an hon-est-but-curious aggregation server that honestly completes data aggregation jobs as pre-defined but tries to infer sensitive information from the uploaded models.”
BackgroundMethodInfluentialFederated Graph Neural Networks: Overview, Techniques, and Challenges
Liu, R., Xing, P., Deng, Z., Li, A., Guan, C., & Yu, H. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3360429 · Cited by 178
“To further protect the privacy, secure aggregation [91] is applied to the relation embedding before being uploaded to the server, and the relation table in the server is obtained via private set union (PSU).”
Ali, M., Karimipour, H., & Tariq, M. (2021) · Computers & Security · doi:10.1016/j.cose.2021.102355 · Cited by 177
From federated learning to federated neural architecture search: a survey
Zhu, H., Zhang, H., & Jin, Y. (2021) · Complex & Intelligent Systems · doi:10.1007/s40747-020-00247-z · Cited by 177
FedCoin: A Peer-to-Peer Payment System for Federated Learning
Liu, Y., Ai, Z., Sun, S., Zhang, S., Liu, Z., & Yu, H. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-63076-8_9 · Cited by 177
Federated Learning: A signal processing perspective
Gafni, T., Shlezinger, N., Cohen, K., Eldar, Y. C., & Poor, H. V. (2022) · IEEE Signal Processing Magazine · doi:10.1109/msp.2021.3125282 · Cited by 176
“, [34], though with additional communication overhead and possible limited compliance to the presence of stragglers and limited device participation.”
BackgroundMaking AI Forget You: Data Deletion in Machine Learning
Ginart, A., Guan, M. Y., Valiant, G., & Zou, J. (2019) · arXiv · doi:10.48550/arxiv.1907.05012 · Cited by 175
“cryptography [63, 16, 14, 13, 62, 31], and differential privacy [30, 21, 20, 64, 1] — do not lead to efficient data deletion, but rather attempt to make data private or non-identifiable.”
BackgroundSource-free unsupervised domain adaptation: A survey
Fang, Y., Yap, P.-T., Lin, W., Zhu, H., & Liu, M. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106230 · Cited by 175
Do Gradient Inversion Attacks Make Federated Learning Unsafe?
Hatamizadeh, A., Yin, H., Molchanov, P., Myronenko, A., Li, W., Dogra, P., Feng, A., Flores, M. G., Kautz, J., Xu, D., et al. (2023) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2023.3239391 · Cited by 125
“In our attack scenario, we consider the server as “honest but curious” [16].”
BackgroundHu, R., Guo, Y., & Gong, Y. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3343288 · Cited by 178
“In the typical setting of FL with a single server, secure aggregation is achieved by adding random mask vectors sampled over a finite group on local model updates before sending them out [18].”
BackgroundMethodFedOpt: Towards Communication Efficiency and Privacy Preservation in Federated Learning
Asad, M., Moustafa, A., & Ito, T. (2020) · Applied Sciences · doi:10.3390/app10082864 · Cited by 174
“Specifically, PSA and XGB utilise collaborative training in order to resist collusion among adversaries, but both approaches do not guarantee communication efficiency.”
MethodFedNLP: Benchmarking Federated Learning Methods for Natural Language Processing Tasks
Lin, B. Y., He, C., Ze, Z., Wang, H., Hua, Y., Dupuy, C., Gupta, R., Soltanolkotabi, M., Ren, X., & Avestimehr, S. (2022) · Findings of the Association for Computational Linguistics: NAACL · doi:10.18653/v1/2022.findings-naacl.13 · Cited by 174
“Aggregation (SA) FedNLP supports state-of-the-art SA algorithms LightSecAgg , SecAgg (Bonawitz et al., 2017), and SecAgg+ (Bell et al., 2020).”
BackgroundInfluentialLabel Leakage and Protection in Two-party Split LearningPreprint
Li, O., Sun, J., Xin, Y., Gao, W., Zhang, H., Xie, J., Smith, V., & Wang, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.08504 · Cited by 173
“…to protect communication privacy in FL generally fall into three categories: 1) cryptographic methods such as secure multi-party computation [e.g., Bonawitz et al., 2017]; 2) system-based methods including trusted execution environments [Subramanyan et al., 2017]; and 3) perturbation methods that…”
BackgroundFederated Learning of Gboard Language Models with Differential Privacy
Xu, Z., Zhang, Y., Andrew, G., Choquette, C., Kairouz, P., Mcmahan, B., Rosenstock, J., & Zhang, Y. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track) · doi:10.18653/v1/2023.acl-industry.60 · Cited by 54
Self-citationEfficient Dropout-Resilient Aggregation for Privacy-Preserving Machine Learning
Liu, Z., Guo, J., Lam, K.-Y., & Zhao, J. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3163592 · Cited by 137
“We notice that in the SecAgg scheme [24], the communication cost can be further reduced by having the clients agree on common seeds or keys.”
BackgroundMethodInfluentialAI augmented Edge and Fog computing: Trends and challenges
Tuli, S., Mirhakimi, F., Pallewatta, S., Zawad, S., Casale, G., Javadi, B., Yan, F., Buyya, R., & Jennings, N. R. (2023) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2023.103648 · Cited by 174
Sun, J., Li, A., Wang, B., Yang, H., Li, H., & Chen, Y. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00919 · Cited by 170
Source Inference Attacks in Federated Learning
Hu, H., Salcic, Z. A., Sun, L., Dobbie, G., & Zhang, X. (2021) · Proceedings · doi:10.1109/icdm51629.2021.00129 · Cited by 172
LSFL: A Lightweight and Secure Federated Learning Scheme for Edge Computing
Zhang, Z., Wu, L., Ma, C., Li, J., Wang, J., Wang, Q., & Yu, S. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3221899 · Cited by 170
Privacy-preserving Heterogeneous Federated Transfer Learning
Gao, D., Liu, Y., Huang, A., Ju, C., Yu, H., & Yang, Q. (2019) · 2019 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata47090.2019.9005992 · Cited by 118
“It has the benefits of both distributed learning [29] and data privacy and confidentiality [3].”
BackgroundSoK: Secure Aggregation Based on Cryptographic Schemes for Federated Learning
Mansouri, M., Önen, M., Jaballah, W. B., & Conti, M. (2023) · Proceedings on Privacy Enhancing Technologies · doi:10.56553/popets-2023-0009 · Cited by 167
How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacy
Ponomareva, N., Hazimeh, H., Kurakin, A., Xu, Z., Denison, C., McMahan, H. B., Vassilvitskii, S., Chien, S., & Thakurta, A. G. (2023) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.14649 · Cited by 166
“This applies even to someone with access to the server under the security assumptions necessary for the private aggregation protocol, which could be provided cryptographically, e.g. via Secure Aggregation (Bonawitz et al., 2017), or via hardware trusted execution environments (TEEs).”
BackgroundSelf-citationOn the Byzantine Robustness of Clustered Federated Learning
Sattler, F., Muller, K.-R., Wiegand, T., & Samek, W. (2020) · ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp40776.2020.9054676 · Cited by 166
“Federated Learning is a distributed training framework, which allows multiple clients (typically mobile or IoT devices) to jointly train a single deep learning model on their combined data in a communication-efficient way, without requiring any of the participants to reveal their private training data to a centralized…”
BackgroundDu, J., Jiang, B., Jiang, C., Shi, Y., & Han, Z. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3242727 · Cited by 167
“Different from the transmissions via wireless channels in federated learning, AirComp enables a “one-shot” data aggregation from concurrent transmissions to harness the interferences and the server receives the superimposed signals to be reconstructed, which is also aligned with the concept of secure aggregation [18].”
MethodHandling privacy-sensitive medical data with federated learning: Challenges and future directions
Aouedi, O., Sacco, A., Piamrat, K., & Marchetto, G. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3185673 · Cited by 165
“These approaches often provide privacy at the cost of reduced model performance or system efficiency [62].”
InfluentialNon-Functional Requirements for Machine Learning: Challenges and New Directions
Horkoff, J. (2019) · IEEE International Requirements Engineering Conference (RE) · doi:10.1109/re.2019.00050 · Cited by 167
“introduce a method to preserve privacy in ML, focusing on keeping the overhead in terms of runtime and communication low [5].”
BackgroundMethodSiloed Federated Learning for Multi-centric Histopathology Datasets
Andreux, M., du Terrail, J. O., Béguier, C., & Tramel, E. W. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-60548-3_13 · Cited by 166
DP-Forward: Fine-tuning and Inference on Language Models with Differential Privacy in Forward Pass
Du, M., Yue, X., Chow, S. S. M., Wang, T., Huang, C., & Sun, H. (2023) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3576915.3616592 · Cited by 167
“…𝑓 ( 𝑖 − 1 ) (·) securely: i) We can assume an extra trusted party (as in DP-SGD), but it becomes central DP. ii) Users can first derive the gradients for the layers inside 𝑓 ( 𝑖 − 1 ) (·) locally on their 𝑋 and then resort to secure aggregation [9] for global updates at the service provider.”
BackgroundPrivacy-Preserving Federated Deep Learning with Irregular Users
Xu, G., Li, H., Zhang, Y., Xu, S., Ning, J., & Deng, R. (2020) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2020.3005909 · Cited by 164
Two-Phase Multi-Party Computation Enabled Privacy-Preserving Federated Learning
Kanagavelu, R., Li, Z., Samsudin, J., Yang, Y., Yang, F., Mong Goh, R. S., Cheah, M., Wiwatphonthana, P., Akkarajitsakul, K., & Wang, S. (2020) · 2020 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID) · doi:10.1109/ccgrid49817.2020.00-52 · Cited by 105
“…it is concluded that in order to ensure the confidentiality of all the information and prevent the indirect data leakage, the models parameters need to be securely encrypted [10] against any possibility of reverse engineering and third party attacks against federated learning model training.”
BackgroundImproving Semi-supervised Federated Learning by Reducing the Gradient Diversity of Models
Zhang, Z., Yang, Y., Yao, Z., Yan, Y., Gonzalez, J. E., Ramchandran, K., & Mahoney, M. W. (2021) · 2021 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata52589.2021.9671693 · Cited by 83
“In addition to the challenge of the non-iidness of the data distribution and the need for local ground truth labels, communication efficiency is another critical problem in FL [37], [42]–[45].”
BackgroundVeriFi: Towards Verifiable Federated Unlearning
Gao, X., Ma, X., Wang, J., Sun, Y., Li, B., Ji, S., Cheng, P., & Chen, J. (2024) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2024.3382321 · Cited by 167
“Furthermore, we specially introduce secure aggregation scheme [21] to safely update the global model without disclosure of individual updates.”
Incentive Mechanism for Horizontal Federated Learning Based on Reputation and Reverse Auction
Zhang, J., Wu, Y., & Pan, R. (2021) · Proceedings of the Web Conference 2021 · doi:10.1145/3442381.3449888 · Cited by 162
“To preserve data privacy, Google proposed a model security aggregation method [4], and Mandal et al.”
MethodFedGraph: Federated Graph Learning With Intelligent Sampling
Chen, F., Li, P., Miyazaki, T., & Wu, C. (2021) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3125565 · Cited by 165
“This is a typical threat model that has been widely used by current federated learning research [16], [28], [29].”
MethodAggregation Service for Federated Learning: An Efficient, Secure, and More Resilient Realization
Zheng, Y., Lai, S., Liu, Y., Yuan, X., Yi, X., & Wang, C. (2023) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2022.3146448 · Cited by 161
“We note that the design of [7] has a double-masking mechanism, which is to prevent the server from learning the data of users who are too late in sending their masked vectors and are assumed to drop-out by the server (so the server recovers their secret keys).”
BackgroundResultInfluentialManipulation Attacks in Local Differential Privacy
Cheu, A., Smith, A., & Ullman, J. (2021) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp40001.2021.00001 · Cited by 160
Awan, S., Li, F., Luo, B., & Liu, M. (2019) · Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3319535.3363256 · Cited by 125
“To address this problem, several secure aggregation algorithms for FL have been proposed recently [2, 3], leveraging secret sharing and differential privacy techniques.”
BackgroundMethodBillion-scale federated learning on mobile clients: a submodel design with tunable privacy
Niu, C., Wu, F., Tang, S., Hua, L., Jia, R., Lv, C., Wu, Z., & Chen, G. (2020) · ACM Annual International Conference on Mobile Computing and Networking (MobiCom) · doi:10.1145/3372224.3419188 · Cited by 162
“[8] designed a secure aggregation protocol in both honest-but-curious and actively adversarial settings, which will be used as a building block of our SFSL.”
BackgroundMethodInfluentialMachine Learning Meets Communication Networks: Current Trends and Future Challenges
Ahmad, I., Shahabuddin, S., Malik, H. H., Harjula, E., Leppänen, T., Lovén, L., Anttonen, A., Sodhro, A. H., Alam, M. M., Juntti, M. J., et al. (2020) · IEEE Access · doi:10.1109/access.2020.3041765 · Cited by 162
“One of the main objectives of the protocol [339] is the privacy preservation of sensitive data using ML in multi-party computation systems.”
BackgroundPrivacy-Preserving Asynchronous Federated Learning Mechanism for Edge Network Computing
Lu, X., Liao, Y., Lio, P., & Hui, P. (2020) · IEEE Access · doi:10.1109/access.2020.2978082 · Cited by 159
“Privacy protection in federated learning, such as security aggregation [18], requires synchronization operations at the device level, so it essentially belongs to the category of synchronous training.”
BackgroundDeep Learning on Mobile and Embedded Devices: State-of-the-art, Challenges, and Future Directions
Chen, Y., Zheng, B., Zhang, Z., Wang, Q., Shen, C., & Zhang, Q. (2021) · ACM Computing Surveys · doi:10.1145/3398209 · Cited by 159
“Google designs a secure aggregation protocol to avoid revealing any individual’s update to the server based on secret sharing [13].”
BackgroundBlanco-Justicia, A., Domingo-Ferrer, J., Martínez, S., Sánchez, D., Flanagan, A., & Tan, K. E. (2021) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2021.104468 · Cited by 159
The Value of Collaboration in Convex Machine Learning with Differential Privacy
Wu, N., Farokhi, F., Smith, D., & Kaafar, M. A. (2020) · 2020 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp40000.2020.00025 · Cited by 122
“Training ML models using encrypted data was discussed in [10]–[14].”
MethodFederated learning for drone authentication
Yazdinejad, A., Parizi, R. M., Dehghantanha, A., & Karimipour, H. (2021) · Ad Hoc Networks · doi:10.1016/j.adhoc.2021.102574 · Cited by 158
Applications of Federated Learning; Taxonomy, Challenges, and Research Trends
Shaheen, M., Farooq, M. S., Umer, T., & Kim, B.-S. (2022) · Electronics · doi:10.3390/electronics11040670 · Cited by 157
“[23] Secure aggregation FL Practicality enhancement CM Privacy guarantees State-of-the-art algorithm for aggregation of data.”
BackgroundGradient-Leakage Resilient Federated Learning
Wei, W., Liu, L., Wu, Y., Su, G., & Iyengar, A. (2021) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs51616.2021.00081 · Cited by 157
Stevens, R. L., USDOE, Taylor, V., Nichols, J., Maccabe, A. B. ‘., Yelick, K., & Brown, D. L. (2020) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · doi:10.2172/1604756 · Cited by 159
VerSA: Verifiable Secure Aggregation for Cross-Device Federated Learning
Hahn, C., Kim, H.-D., Kim, M., & Hur, J. (2021) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3126323 · Cited by 156
“The SA [37] is a baseline privacy-preserving model aggregation protocol on which V ER SA is built; thus, we demonstrate the cost of achieving verifiability over SA by comparing it with SA.”
BackgroundMethodInfluentialPrivacy and Fairness in Federated Learning: On the Perspective of Tradeoff
Chen, H., Zhu, T., Zhang, T., Zhou, W., & Yu, P. S. (2023) · ACM Computing Surveys · doi:10.1145/3606017 · Cited by 159
“Bonawitz et al. [15, 137] proposed a secure aggregation method for FL based on 𝑡 -out-of- 𝑛 secret sharing.”
MethodShielding Collaborative Learning: Mitigating Poisoning Attacks through Client-Side Detection
Zhao, L., Hu, S., Wang, Q., Jiang, J., Chao, S., Luo, X., & Hu, P. (2020) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2020.2986205 · Cited by 127
The role of cross-silo federated learning in facilitating data sharing in the agri-food sector
Durrant, A., Marković, M., Matthews, D., May, D., Enright, J. A., & Leontidis, G. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106648 · Cited by 162
“Although many of the privacy preservation schemes explored defend from such public attacks, while secure communication pipelines like secure aggregation [21] can help reduce such risks.”
Federated Learning for Privacy-Preserving: A Review of PII Data Analysis in Fintech
Dash, B., Sharma, P., & Ali, A. (2022) · International Journal of Software Engineering & Applications · doi:10.5121/ijsea.2022.13401 · Cited by 155
“The idea allowed most scientists to be amazed by the capabilities of the entire federated learning system [3].”
BackgroundBDFL: A Byzantine-Fault-Tolerance Decentralized Federated Learning Method for Autonomous Vehicle
Chen, J.-H., Chen, M.-R., Zeng, G.-Q., & Weng, J.-S. (2021) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2021.3102121 · Cited by 122
“1) The Untrusted Leader: In the aggregation protocol of FL [30], the honest-but-curious secure Multi-Party Computation (MPC) server is required.”
BackgroundMethodNaghib, A., Gharehchopogh, F. S., & Zamanifar, A. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-024-11101-w · Cited by 165
EIFFeL: Ensuring Integrity for Federated Learning
Chowdhury, A. R., Guo, C., Jha, S., & van der Maaten, L. J. P. (2022) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3548606.3560611 · Cited by 156
“Prior work has addressed the problem of (non-Byzantine) secure aggregation in FL [3, 9, 15, 66].”
BackgroundMethodIntrusion Detection Based on Federated Learning: A Systematic Review
Hernández-Ramos, J. L., Καρόπουλος, Γ., Chatzoglou, E., Kouliaridis, V., Campos, E. M., González-Vidal, A., & Kambourakis, G. (2025) · ACM Computing Surveys · doi:10.1145/3731596 · Cited by 155
Membership Privacy for Machine Learning Models Through Knowledge Transfer
Shejwalkar, V., & Houmansadr, A. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i11.17150 · Cited by 157
Yazdinejad, A., Dehghantanha, A., Srivastava, G., Karimipour, H., & Parizi, R. M. (2024) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2024.103088 · Cited by 155
AriaNN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharing
Ryffel, T., Tholoniat, P., Pointcheval, D., & Bach, F. (2022) · Proceedings on Privacy Enhancing Technologies · doi:10.2478/popets-2022-0015 · Cited by 96
An Overview of Privacy in Machine LearningPreprint
Cristofaro, E. D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.08679 · Cited by 153
“Then, Bonawitz et al. [5] use secure multi-party aggregation techniques, tailored for federated learning, to let participants encrypt their updates so that the central parameter server only recovers the sum of the updates.”
BackgroundEPPDA: An Efficient Privacy-Preserving Data Aggregation Federated Learning Scheme
Song, J., Wang, W., Gadekallu, T. R., Cao, J., & Liu, Y. (2023) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3153519 · Cited by 150
Zhan, Y., Li, P., & Guo, S. (2020) · 2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps47924.2020.00033 · Cited by 124
“[42] have proposed a communication-efficient and failurerobust protocol for secure aggregation of high-dimensional data.”
BackgroundTrustworthy federated learning: privacy, security, and beyond
Chen, C., Liu, J., Tan, H., Li, X., Wang, K. I., Li, P., Sakurai, K., & Dou, D. (2024) · Knowledge and Information Systems · doi:10.1007/s10115-024-02285-2 · Cited by 152
“SS, on the other hand, involves the dissemination of key shares across multiple entities, enabling decryption only by authorized coalitions [18].”
Federated Reconstruction: Partially Local Federated Learning
Singhal, K., Sidahmed, H., Garrett, Z., Wu, S., Rush, K., & Prakash, S. (2021) · neural information processing systems · doi:10.48550/arxiv.2102.03448 · Cited by 151
“Thus, in the most privacy-sensitive applications, F ED R ECON may be augmented with differential privacy [17] or secure aggregation [8] applied to global parameter updates, to provide provable guarantees about the information the server receives.”
BackgroundWAFFLE: Watermarking in Federated Learning
Tekgul, B. G. A., Xia, Y., Marchal, S., & Asokan, N. (2021) · IEEE Symposium on Reliable Distributed Systems (SRDS) · doi:10.1109/srds53918.2021.00038 · Cited by 150
Ma, Y., Woods, J., Angel, S., Polychroniadou, A., & Rabin, T. (2023) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp46215.2023.10179434 · Cited by 153
FedGKD: Toward Heterogeneous Federated Learning via Global Knowledge Distillation
Yao, D., Pan, W., Dai, Y., Wan, Y., Ding, X., Chen, Y., Jin, H., Xu, Z., & Sun, L. (2023) · IEEE Transactions on Computers · doi:10.1109/tc.2023.3315066 · Cited by 152
“However, as discussed in [15], [26], the partial global sharing of local information violates privacy, and using globally-shared auxiliary data should be cautiously generated or collected.”
BackgroundArchitectural patterns for the design of federated learning systems
Lo, S. K., Lu, Q., Zhu, L., Paik, H.-Y., Xu, X., & Wang, C. (2022) · Journal of Systems and Software · doi:10.1016/j.jss.2022.111357 · Cited by 149
Understanding Clipping for Federated Learning: Convergence and Client-Level Differential PrivacyPreprint
Zhang, X., Chen, X., Hong, M., Wu, Z. S., & Yi, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.13673 · Cited by 151
“During the communication, we assume that the attacker can only observe the aggregated update i ∈P t ˜∆ x ti , and this can be guaranteed by using secure aggregation (Bonawitz et al., 2017) or assuming secure uplinks of the clients.”
MethodChallenges, Applications and Design Aspects of Federated Learning: A Survey
Rahman, K. M. J., Ahmed, F., Akhter, N., Hasan, M., Amin, R., Aziz, K. E., Islam, A. K. M. M., Mukta, M. S. H., & Islam, A. K. M. N. (2021) · IEEE Access · doi:10.1109/access.2021.3111118 · Cited by 147
Secure and Efficient Federated Transfer Learning
Sharma, S., Xing, C., Liu, Y., & Kang, Y. (2019) · 2019 IEEE International Conference on Big Data (Big Data) · doi:10.1109/bigdata47090.2019.9006280 · Cited by 110
“For security against such attacks, our techniques can be merged with secure aggregation techniques introduced in [25].”
MethodFederated Latent Dirichlet Allocation: A Local Differential Privacy Based Framework
Wang, Y., Tong, Y., & Shi, D. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i04.6096 · Cited by 116
“Some solutions consider collaborative machine learning under the secure multiparty computation (SMC) framework (Bonawitz et al. 2017; Cheng et al. 2019) and apply encryption protocols to satisfy the privacy requirements.”
BackgroundFederated Model Distillation with Noise-Free Differential Privacy
Sun, L., & Lyu, L. (2021) · Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/216 · Cited by 114
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.