Byzantine-Resilient Secure Federated Learning
Open accessThe DOI matches its Crossref record; 47 of the 48 references it cites were found in a registry or an index.
Abstract
Secure federated learning is a privacy-preserving framework to improve machine learning models by training over large volumes of data collected by mobile users. This is achieved through an iterative process where, at each iteration, users update a global model using their local datasets. Each user then masks its local update via random keys, and the masked models are aggregated at a central server to compute the global model for the next iteration. As the local updates are protected by random masks, the server cannot observe their true values. This presents a major challenge for the resilience of the model against adversarial (Byzantine) users, who can manipulate the global model by modifying their local updates or datasets. Towards addressing this challenge, this paper presents the first single-server Byzantine-resilient secure aggregation framework (BREA) for secure federated learning. BREA is based on an integrated stochastic quantization, verifiable outlier detection, and secure model aggregation approach to guarantee Byzantine-resilience, privacy, and convergence simultaneously. We provide theoretical convergence and privacy guarantees and characterize the fundamental trade-offs in terms of the network size, user dropouts, and privacy protection. Our experiments demonstrate convergence in the presence of Byzantine users, and comparable accuracy to conventional federated learning benchmarks.
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
“Bayesian neural networks [419] have been proposed and shown to be far more robust to over-fitting, and can easily learn from small datasets.”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [419]
In the citing paper
“To simultaneously preserve privacy for individual users while tolerating Byzantine adversaries, a Byzantine-resilient secure aggregation framework was developed in [171] to detect adversarial models without the knowledge of individual local models, as they are masked for privacy guarantees.”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [171]
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
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
Shamir, A. (1979) · Communications of the ACM · doi:10.1145/359168.359176 · Cited by 13,756 · cited 5×
“Our framework leverages Feldman’s verifiable secret sharing protocol from [17], which combines Shamir’s secret sharing [18] with homomorphic encryption.”
BackgroundMethodInfluentialAdvances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
Federated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210
Practical Secure Aggregation for Privacy-Preserving Machine Learning
Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., Ramage, D., Segal, A., & Seth, K. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3133982 · Cited by 3,956
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
Differential Privacy: A Survey of Results
Dwork, C. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-79228-4_1 · Cited by 3,542 · cited 1×
“Although our focus is not on differential-privacy, our approach may in principle be combined with differential privacy techniques [27], which is an interesting future direction.”
Background
Research works printed but not linked: 3
- bottou, 1998, Online learning and stochastic approximations, Online Learning in Neural Networks, 17, 142
- mcmahan, 2017, Communication-efficient learning of deep networks from decentralized data, Proc 20th Int Conf Artif Intell Statist, 54, 1273
- lecun, 2010, MNIST Handwritten Digit Database
Cited by
Advances and Open Problems in Federated Learning
Kairouz, P., & McMahan, H. B. (2021) · Foundations and Trends in Machine Learning · doi:10.1561/2200000083 · Cited by 7,772
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection
Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., & He, B. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3124599 · Cited by 1,407
“Inspired by robust aggregation in Byzantine tolerant distributed learning [19], [157], where the number of tolerated Byzantine adversaries is used to evaluate the robustness, we”
MethodEdge artificial intelligence for 6g: vision, enabling technologies, and applications
Letaief, K. B., Shi, Y., Lu, J., & Lu, J. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126076 · Cited by 829
“To simultaneously preserve privacy for individual users while tolerating Byzantine adversaries, a Byzantine-resilient secure aggregation framework was developed in [171] to detect adversarial models without the knowledge of individual local models, as they are masked for privacy guarantees.”
MethodPrivacy-Enhanced Federated Learning Against Poisoning Adversaries
Liu, X., Li, H., Xu, G., Chen, Z., Huang, X., & Lu, R. (2021) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2021.3108434 · Cited by 351
“The closest to our method we noticed is [22], as it focuses on making the current distance-based outlier removal mechanism compatible with secure aggregation utilizing SMC.”
MethodResultFedML: A Research Library and Benchmark for Federated Machine Learning
He, C., Li, S., So, J., Zeng, X., Zhang, M., Wang, H., Wang, X., Vepakomma, P., Singh, A., Qiu, H., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.13518 · Cited by 314
“To tackle the challenge of trustworthiness, existing research focuses on developing new defense techniques for adversarial attacks to make FL robust [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 24, 27, 28], and proposing methods such as differential privacy (DP) and secure multiparty computation (SMPC) to protect…”
BackgroundSelf-citationBlockchain-Based Federated Learning for Securing Internet of Things: A Comprehensive Survey
Issa, W., Moustafa, N., Turnbull, B., Sohrabi, N., & Tari, Z. (2023) · ACM Computing Surveys · doi:10.1145/3560816 · Cited by 319
“[137] presented a Byzantine resilience approach.”
Turbo-Aggregate: Breaking the Quadratic Aggregation Barrier in Secure Federated Learning
So, J., Guler, B., & Avestimehr, A. S. (2021) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2021.3054610 · Cited by 318
“approach is combining secure aggregation with an outlier detection algorithm as proposed in [52], which has a communication cost of O(N2) that limits its scalability to large federated learning systems.”
MethodSelf-citationGroup Knowledge Transfer: Federated Learning of Large CNNs at the Edge
He, C., Annavaram, M., & Avestimehr, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.14513 · Cited by 177
Duan, Q., Huang, J., Hu, S., Deng, R., Lu, Z., & Yu, S. (2023) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2023.3316615 · Cited by 159
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 169
“So et al. [30] also proposed a Byzantine-resistant secure aggregation framework (BREA) for secure FL.”
BackgroundFastSecAgg: Scalable Secure Aggregation for Privacy-Preserving Federated Learning
Kadhe, S., Rajaraman, N., Koyluoglu, O. O., & Ramchandran, K. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.11248 · Cited by 108
“Here, we review the basics of the finite field Fourier transform, for details, see, e.g., [36].”
MethodTrustworthy Federated Learning via Blockchain
Yang, Z., Shi, Y., Zhou, Y., Wang, Z., & Yang, K. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3201117 · Cited by 122
“Although FL guarantees the privacy protection of edge devices, the global model can be attacked by malicious edge devices (e.g., model poisoning or adversarial attacks) [12].”
BackgroundFedNLP: 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
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 188
“These protocols can be divided based on whether they operate in the single-aggregator model where a central server facilitates gradient aggregation (RoFL [34], EIFFeL [40], [88, 102]), or if they distribute the trust of the central aggregator into two servers hosted behind separate trust domains (Prio [41], Prio+ [13]…”
BackgroundKhraisat, A., Alazab, A., Singh, S., Jan, T., & Gomez, A. J. (2024) · ACM Computing Surveys · doi:10.1145/3687124 · Cited by 189
“The server aggregates the received models/updates to build the global model [33].”
Federated Graph Neural Network for Fast Anomaly Detection in Controller Area Networks
Zhang, H., Zeng, K., & Lin, S. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3240291 · Cited by 125
“Currently, Byzantine-tolerant distributed learning [54], [55], [56] is widely considered as one potential solution.”
BackgroundPrivacy-Preserving Federated Learning With Malicious Clients and Honest-but-Curious Servers
Le, J., Zhang, D., Lei, X., Jiao, L., Zeng, K., & Liao, X. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3295949 · Cited by 125
“In recent years, numerous researchers have studied the aforementioned security issues of FL and proposed various defense solutions such as the Secure Multiparty Computation (SMC)-based schemes [13], the Homomorphic Encryption (HE)-based schemes [14], the Differential Privacy (DP)-based schemes [15], [16], [17], the by…”
BackgroundMethodSoK: 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
Advanced Deep Learning Models for 6G: Overview, Opportunities, and Challenges
Jiao, L., Shao, Y., Sun, L., Liu, F., Yang, S., Ma, W., Li, L., Liu, X., Hou, B., Zhang, X., et al. (2024) · IEEE Access · doi:10.1109/access.2024.3418900 · Cited by 148
Federated Learning With Fair Incentives and Robust Aggregation for UAV-Aided Crowdsensing
Wang, Y., Su, Z., Luan, T. H., Li, R., & Zhang, K. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3138928 · Cited by 104
“Thus, the FL server can apply robust estimation rules, such as geometric median [59], marginal trimmed mean [60], and Krum [54], to achieve noise-tolerant learning.”
BackgroundMethodSecurity and Privacy Threats to Federated Learning: Issues, Methods, and Challenges
Zhang, J., Zhu, H., Wang, F., Zhao, J., Xu, Q., & Li, H. (2022) · Security and Communication Networks · doi:10.1155/2022/2886795 · Cited by 145
“As a result, Pairwise Masking can thwart model updates from being revealed when the adversary aggregator reconstructs the participant’s perturbation [85].”
BackgroundDifferentially Private Byzantine-Robust Federated Learning
Ma, X., Sun, X., Wu, Y., Liu, Z., Chen, X., & Dong, C. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2022.3167434 · Cited by 109
“In the study of federated learning, few schemes [32], [33], [34], [35] address both privacy and robustness.”
BackgroundPrivate Retrieval, Computing, and Learning: Recent Progress and Future Challenges
Ulukuş, Ş., Avestimehr, S., Gastpar, M., Jafar, S. A., Tandon, R., & Tian, C. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3142358 · Cited by 119
Tariq, A., Serhani, M. A., Sallabi, F., Barka, E. S., Qayyum, T., Khater, H. M., & Shuaib, K. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3438264 · Cited by 155
Trustworthy 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 151
“These range from poisoning offenses [9] and Byzantine stratagems [35, 194], to intricate inference attacks [190].”
Chain FL: Decentralized Federated Machine Learning via Blockchain
Korkmaz, C., Kocas, H. E., Uysal, A., Masry, A., Ozkasap, O., & Akgun, B. (2020) · 2020 Second International Conference on Blockchain Computing and Applications (BCCA) · doi:10.1109/bcca50787.2020.9274451 · Cited by 84
“Poor quality updates resulting from poor data quality, faulty updates and malicious updates can all be prevented either by using a centralized test dataset to check update performance [12], [13], or by using decentralized test datasets other parties have [14].”
BackgroundEIFFeL: 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
“Additionally, we compare EIFFeL’s performance with BREA [80], which is a Byzantine-robust secure aggregator.”
BackgroundMethodInfluentialHeteroSAg: Secure Aggregation With Heterogeneous Quantization in Federated Learning
Elkordy, A. R., & Avestimehr, A. S. (2022) · IEEE Transactions on Communications · doi:10.1109/tcomm.2022.3151126 · Cited by 118
“In recent work, Byzantine-robust secure aggregation algorithms have been proposed [26,27].”
BackgroundSelf-citationImproving Federated Learning With Quality-Aware User Incentive and Auto-Weighted Model Aggregation
Deng, Y., Lyu, F., Ren, J., Chen, Y., Yang, P., Zhou, Y., & Zhang, Y. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2022.3195207 · Cited by 89
“Recently, there has been extensive research attention dedicated to the performance optimization of federated learning from the aspects of communication [35], [36], [37], privacy [29], robustness [38], personalization [39], and incentive [40], [41] [38].”
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
Privacy-preserving Byzantine-robust federated learning
Ma, X., Zhou, Y., Wang, L., & Miao, M. (2021) · Computer Standards & Interfaces · doi:10.1016/j.csi.2021.103561 · Cited by 83
Information Theoretic Secure Aggregation With User Dropouts
Zhao, Y., & Sun, H. (2022) · IEEE Transactions on Information Theory · doi:10.1109/tit.2022.3192874 · Cited by 97
Efficient, Private and Robust Federated Learning
Hao, M., Li, H., Xu, G., Chen, H., & Zhang, T. (2021) · Annual Computer Security Applications Conference · doi:10.1145/3485832.3488014 · Cited by 84
“Different from the Krum-based works [24] [43] that utilize the Euclidean distance, the cosine similarity measurement additionally takes into account the direction of parties’ updates.”
BackgroundMethodInfluentialSVeriFL: Successive verifiable federated learning with privacy-preserving
Gao, H., He, N., & Gao, T. (2022) · Information Sciences · doi:10.1016/j.ins.2022.11.124 · Cited by 72
SafeFL: MPC-friendly Framework for Private and Robust Federated Learning
Gehlhar, T., Marx, F., Schneider, T. H., Suresh, A., Wehrle, T., & Yalame, H. (2023) · IEEE Security and Privacy Workshops (SPW) · doi:10.1109/spw59333.2023.00012 · Cited by 87
“FedAvg [59] – ✗ ✗ ✗ ✗ ✗ ✓ ✓ ✗ Krum [10] – ✗ ✓ ✗ ✗ ✗ ✗ ✓ ✓ SecAgg [15] Masking ✓ ✗ ✓ ✗ ✗ ✓ ✗ ✗ PPDL [71] HE ✓ ✗ ✗ ✗ ✗ ✓ ✓ ✗ Median [87] – ✗ ✓ ✗ ✗ ✗ ✗ ✓ ✗ SecAgg+ [8] Masking ✓ ✗ ✓ ✗ ✓ ✓ ✗ ✗ BatchCrypt [88] HE ✓ ✗ ✓ ✗ ✗ ✓ ✓ ✓ FoolsGold [37] – ✗ ✓ ✗ ✗ ✗ ✗ ✓ ✓ SAFELearn [35] MPC ✓ ✗ ✗ ✓ ✓ ✓ ✓ ✓ BREA [77] Masking ✓ ✓ ✗ ✗ ✗…”
FedMes: Speeding Up Federated Learning With Multiple Edge Servers
Han, D.-J., Choi, M., Park, J., & Moon, J. J. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118422 · Cited by 65
“The convergence behavior of FL schemes are studied in [6], [12], and security/privacy issues of FL are addressed in [19].”
BackgroundLiu, W., Lin, H., Wang, X., Hu, J. W., Kaddoum, G., Piran, M. J., & Alamri, A. M. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3081606 · Cited by 67
“Although the federated learning can protect privacy better than the traditional ML, the server cannot verify the security and availability of local models of the participants [9]–[11].”
BackgroundFederated 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 68
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…”
BackgroundHouda, Z. A. E., Moudoud, H., Brik, B., & Khoukhi, L. (2023) · IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) · doi:10.1109/infocomwkshps57453.2023.10225769 · Cited by 61
ADFL: A Poisoning Attack Defense Framework for Horizontal Federated Learning
Guo, J., Li, H., Huang, F., Liu, Z., Peng, Y., Li, X., Ma, J., Menon, V. G., & Igorevich, K. K. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3156645 · Cited by 49
TDFL: Truth Discovery Based Byzantine Robust Federated Learning
Xu, C. L., Jia, Y., Zhu, L., Zhang, C., Jin, G., & Sharif, K. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2022.3205714 · Cited by 62
“FedAvg is the basic aggregation scheme for federated learning, similar to prior works [11], [24], we use the FedAvg method as the baseline.”
MethodZheng, J., Tian, H., Ni, W., Ni, W., & Zhang, P. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3188502 · Cited by 47
“The authors of [25] extended the Krum algorithm to improve the resilience to Byzantine attacks by selecting the most plausible set of users for model aggregation.”
BackgroundMethodEnhanced 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
“BREA [44] proposes a single-server framework where each participant secret-shares her local update with all the other participants in the system.”
BackgroundXing, Z., Zhang, Z., Zhang, Z., Li, Z., Li, M., Liu, J., Zhang, Z., Zhao, Y., Sun, Q., Zhu, L., et al. (2025) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2025.3561657 · Cited by 109
“Experiment shows that zPROBE achieves lower computation complexity compared to both BREA [178] and EIFFeL [100] and higher accuracy compared to EIFFeL.”
InfluentialFedCV: A Federated Learning Framework for Diverse Computer Vision TasksPreprint
He, C., Shah, A., Tang, Z., Sivashunmugam, D. F. N., Bhogaraju, K., Shimpi, M., Shen, L., Chu, X., Soltanolkotabi, M., & Avestimehr, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.11066 · Cited by 111
“sonalization [75, 14, 42, 55, 28, 20, 10], system efficiency (communication and computation) [23], robustness to attacks [67, 2, 15, 1, 68, 7, 58, 13, 7, 56, 49], and privacy [4, 16, 48, 53, 45, 62, 60, 57, 71, 61].”
BackgroundPrivacy Preservation for Federated Learning With Robust Aggregation in Edge Computing
Liu, W., Xu, X., Li, D., Qi, L., Dai, F., Dou, W., & Ni, Q. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3229122 · Cited by 50
“Provided that the malicious ESs intentionally change the model update (such as gradients) to be uploaded, the inference performance of the aggregated global model may sharply decline [16].”
BackgroundFedGT: 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 67
“To the best of our knowledge, only the works [25], [26], [27] address resiliency against poisoning attacks in conjunction with secure aggregation.”
InfluentialTrusted Decentralized Federated Learning
Gholami, A., Torkzaban, N., & Baras, J. S. (2022) · IEEE Consumer Communications and Networking Conference · doi:10.1109/ccnc49033.2022.9700624 · Cited by 55
Zhang, Z., Guo, B., Sun, W. Y., Liu, Y., & Yu, Z. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3223944 · Cited by 56
“In order to distinguish malicious clients from non-malicious clients, in future work, we will consider integrating intelligent authorization [1] and adding technology to identify Byzantine models [44] to improve the possibility of our framework being deployed in the real world.”
BackgroundA 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
Asad, M., Moustafa, A., & Aslam, M. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107235 · Cited by 41
Rodríguez-Barroso, N., Martínez-Cámara, E., Luzón, M. V., & Herrera, F. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.108588 · Cited by 38
A Multi-Shuffler Framework to Establish Mutual Confidence for Secure Federated Learning
Zhou, Z., Xu, C., Wang, M., Kuang, X., Zhuang, Y., & Yu, S. (2022) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2022.3215574 · Cited by 43
“For example, [23] use stochastic quantization and verifiable outlier detection to guarantee Byzantine-resilience and privacy simultaneously.”
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
“The schemes [16] employs the Multi-Krum method and [17] designs a k -nearest weight selection method to detect poisoning attack, while preserving the privacy through secret sharing.”
MethodInfluentialReputation-Aware Federated Learning Client Selection Based on Stochastic Integer Programming
Tan, X., Ng, W., Lim, W. Y. B., Xiong, Z.-H., Niyato, D., & Yu, H. (2024) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2022.3191332 · Cited by 36
“In the worst case, a malicious worker node can intentionally launch serious adversarial attacks, such as backdoor attacks [3] or byzantine attacks [29].”
BackgroundFedEdge: Accelerating Edge-Assisted Federated Learning
Wang, K., He, Q., Chen, F., Jin, H., & Yang, Y. J. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583264 · Cited by 44
“In this experiment, Multi-Krum always removes 𝐾 / 3 models, following its applications in many existing studies on federated learning [61, 63], while Adaptive-Krum follows the process presented in Section 4.2 to filter out anomalous models with 𝑐𝑣 = 1 .”
BackgroundMethodInfluentialNi, L., Gong, X., Li, J., Tang, Y. A., Luan, Z., & Zhang, J. (2023) · Information Sciences · doi:10.1016/j.ins.2023.119784 · Cited by 49
Blockchain-Inspired Collaborative Cyber-Attacks Detection for Securing Metaverse
Zainudin, A., Putra, M. A. P., Alief, R. N., Akter, R., Kim, D.-S., & Lee, J.-M. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3364247 · Cited by 35
“Malicious clients can participate in the FL system and carry out Byzantine attacks by sending inaccurately updated models or false information during aggregation, thus compromising the reliability of the model [12], [13].”
BackgroundFedNLP: A Research Platform for Federated Learning in Natural Language Processing.
Lin, B., He, C., Zeng, Z., Wang, H., Huang, Y., Soltanolkotabi, M., Ren, X., & Avestimehr, S. (2021) · Cited by 45
Auto-weighted Robust Federated Learning with Corrupted Data Sources
Li, S., Ngai, E. C. ‐., Ye, F., & Voigt, T. (2022) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3517821 · Cited by 43
Optimizing Federated Learning With Aggregation Strategies: A Comprehensive Survey
Khan, N., Nisar, S., Khan, M. A., Rehman, Y. A. U., Noor, F., & Barb, G. (2025) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2025.3590102 · Cited by 46
“This approach therefore minimizes the number of communications needed, especially in many client settings [29].”
TruFLaaS: Trustworthy Federated Learning as a Service
Mazzocca, C., Romandini, N., Mendula, M., Montanari, R., & Bellavista, P. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3282899 · Cited by 42
“Although the FL paradigm enables tackling some AI-based challenges, such as preserving privacy, the global model can still be the target of different attacks (i.e., model poisoning and inference attack) [16].”
BackgroundSecure 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 59
Trusted Model Aggregation With Zero-Knowledge Proofs in Federated Learning
Ma, R., Hwang, K., and, M. L., & Miao, Y. (2024) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2024.3455762 · Cited by 36
Robust and Secure Federated Learning With Verifiable Differential Privacy
Zhang, C., Weng, J., Weng, J., Zhong, Y., Liu, J.-N., & Deng, C. (2025) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2025.3574745 · Cited by 43
Adaptive Verifiable Coded Computing: Towards Fast, Secure and Private Distributed Machine Learning
Tang, T., Ali, R. E., Hashemi, H., Gangwani, T., Avestimehr, S., & Annavaram, M. (2022) · IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps53621.2022.00067 · Cited by 45
Privacy preserving and secure robust federated learning: A survey
Han, Q., Lu, S., Wang, W., Qu, H., Li, J., & Gao, Y. (2024) · Concurrency and Computation: Practice and Experience · doi:10.1002/cpe.8084 · Cited by 29
Learning from Failures: Secure and Fault-Tolerant Aggregation for Federated Learning
Mansouri, M., Önen, M., & Jaballah, W. B. (2022) · Annual Computer Security Applications Conference · doi:10.1145/3564625.3568135 · Cited by 35
“This scheme has been used as a building block for a significant number of privacy-preserving federated learning solutions [6, 15, 4, 3, 37, 19, 35, 41, 17, 36].”
MethodTrustiness-based hierarchical decentralized federated learning
Li, Y., Wang, X., Sun, R., Xie, X., Ying, S., & Ren, S. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110763 · Cited by 35
Turbo-Aggregate: Breaking the Quadratic Aggregation Barrier in Secure Federated LearningPreprint
So, J., Güler, B., & Avestimehr, A. S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.04156 · Cited by 28
“approach is combining secure aggregation with an outlier detection algorithm as proposed in [52], which has a communication cost of O(N2) that limits its scalability to large federated learning systems.”
MethodSelf-citationDefendFL: A Privacy-Preserving Federated Learning Scheme Against Poisoning Attacks
Liu, J., Li, X., Liu, X., Zhang, H., Miao, Y., & Deng, R. H. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3423397 · Cited by 30
“11(d) shows that the number of users has little impact on the communication cost of the server-side in the works [35], [40] and DefendFL, while has a greater influence on schemes [30], [41].”
InfluentialByzantine-robust federated learning via credibility assessment on non-IID data
Zhai, K., Ren, Q., Wang, J., & Yan, C. (2021) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2022078 · Cited by 36
Mao, Y., Ye, Z., Yuan, X., & Zhong, S. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3416042 · Cited by 33
Byzantine-Resilient Federated Machine Learning via Over-the-Air Computation
Huang, S., Zhou, Y., Wang, T., & Shi, Y. (2021) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops50388.2021.9473694 · Cited by 31
“from the information-the-oretical perspective [22], [23], model with regularization term [24], and adversarial detection [25], [26].”
BackgroundFLAD: Byzantine-Robust Federated Learning Based on Gradient Feature Anomaly Detection
Tang, P., Zhu, X., Qiu, W., Huang, Z. R., Mu, Z., & Li, S. (2025) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2025.3542437 · Cited by 26
A Secure Personalized Federated Learning Algorithm for Autonomous Driving
Fu, Y., Tang, X., Li, C., Yu, F. R., & Cheng, N. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3450726 · Cited by 33
FedRDF: A Robust and Dynamic Aggregation Function Against Poisoning Attacks in Federated Learning
Campos, E. M., Gonzalez-Vidal, A., Hernández-Ramos, J. L., & Skarmeta, A. (2025) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2024.3474484 · Cited by 22
PRoT-FL: A privacy-preserving and robust Training Manager for Federated Learning
Gamiz, I., Regueiro, C., Jacob, E., Lage, Ó., & Higuero, M. (2024) · Information Processing & Management · doi:10.1016/j.ipm.2024.103929 · Cited by 23
Long-Term Privacy-Preserving Aggregation With User-Dynamics for Federated Learning
Liu, Z., Lin, H.-Y., & Liu, Y. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3266919 · Cited by 29
“Considering adversarial FL users with different capabilities, some works propose PPAgg protocols by adopting privacy-preserving techniques under different threat models [47], [48].”
MethodReinforcement Learning-Based Device Scheduling for Renewable Energy-Powered Federated Learning
Cui, Y., Cao, K., & Wei, T. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3210008 · Cited by 23
“We consider a common FL system in an IoT edge network, and the number of participating IoT devices is 40 in this FL system [24].”
MethodStatistics and Data Science for Cybersecurity
Hero, A. O., Kar, S., Moura, J. M. F., Neil, J., Poor, H. V., Turcotte, M., & Xi, B. (2023) · Harvard Data Science Review · doi:10.1162/99608f92.a42024d0 · Cited by 28
zPROBE: Zero Peek Robustness Checks for Federated Learning
Ghodsi, Z., Javaheripi, M., Sheybani, N., Zhang, X., Huang, K., & Koushanfar, F. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00448 · Cited by 36
“Table 6: Runtime complexity of zPROBE vs. prior works BREA [34] and EIFFeL [10].”
BackgroundMethodResultInfluentialA Flexible and Scalable Malicious Secure Aggregation Protocol for Federated Learning
Tang, J., Xu, H., Wang, M., Tang, T., Peng, C., & Liao, H. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3375527 · Cited by 27
“Many works provide input validation [22], [23], [25], [27], [28], [29], [30], [31], [32], [33], [34].”
NSPFL: A Novel Secure and Privacy-Preserving Federated Learning With Data Integrity Auditing
Zhang, Z., & Li, Y. (2024) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2024.3379852 · Cited by 30
“Next, we first analyzed the computational and communica-tional overhead of NSPFL and compared it with other similar schemes [21], [22], [25], [32], [42], where [21], [22], [25] are standard FL schemes, and [32], [42] are the new variants of FL scheme.”
MethodZhang, Z., Wu, L., Jin, J., Wang, E., Liu, B., & Han, Q. (2025) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2025.3528569 · Cited by 25
ROBY: A Byzantine-Robust and Privacy-Preserving Serverless Federated Learning Framework
Tang, X., Li, M., Shen, M., Kang, J., Zhu, L., Liu, Z., Yang, G., Niyato, D. T., & Deng, R. H. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3589066 · Cited by 27
“Besides, Byzantine-resilient FL systems are proposed to ensure the convergence of SGD despite the presence of a minority of Byzantine attackers [44], [45], [46].”
Basil: A Fast and Byzantine-Resilient Approach for Decentralized Training
Elkordy, A. R., Prakash, S., & Avestimehr, S. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3191347 · Cited by 35
Collaborative Byzantine Resilient Federated Learning
Gouissem, A., Abualsaud, K., Yaacoub, E. E., Khattab, T. M., & Guizani, M. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3266347 · Cited by 26
“[21] used a key sharing and participants selection mechanisms based on private pairwise distance calculations.”
MethodA Hierarchical Blockchain-Enabled Secure Aggregation Algorithm for Federated Learning in IoV
Fu, Y., Niu, X., Zhou, L., Cai, X., Yu, F. R., Cheng, N., & Li, C. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3489032 · Cited by 25
“However, due to the instability of the central server in FL, it is highly susceptible to various attacks from malicious vehicles [12].”
Secure distributed estimation under Byzantine attack and manipulation attack
Wan, F., Ma, T., Hua, Y., Liao, B., & Qing, X. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105384 · Cited by 19
Byzantine-Robust Federated Learning with Variance Reduction and Differential Privacy
Zhang, Z., & Hu, R. (2023) · IEEE Conference on Communications and Network Security · doi:10.1109/cns59707.2023.10288938 · Cited by 31
“Several studies [32], [34] have investigated the use of secure multi-party computation (MPC) combined with sophisticated filters to address the challenges posed by Byzantine attack and privacy inference attack at the same time.”
BackgroundBlockchain‐Based Federated Learning Methodologies in Smart Environments for Drone Technology
Chalapathi, M. M. V., Sreenivasulu, K., Jeya, R., Faheem, M., Mohana, R. M., Khan, A. A., & Ramana, K. (2025) · IEEJ Transactions on Electrical and Electronic Engineering · doi:10.1002/tee.70025 · Cited by 22
“For the first time in history, a Byzantine-resistant secure aggregation architecture for secure federated learning, BREA, has been created [13].”
Information Theoretic Secure Aggregation with User Dropouts
Zhao, Y., & Sun, H. (2021) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit45174.2021.9517953 · Cited by 24
Federated Learning Stability Under Byzantine Attacks
Gouissem, A., Abualsaud, K., Yaacoub, E. E., Khattab, T. M., & Guizani, M. (2022) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc51071.2022.9771594 · Cited by 25
“Pair-wise secret key sharing is used in [15] to add an extra layer of privacy, to secure the communications and to make the FL training resilient to Byzantine attacks.”
BackgroundFederated clustered multi-domain learning for health monitoring
Jiang, S., Li, Y., Firouzi, F., & Chakrabarty, K. (2024) · Scientific Reports · doi:10.1038/s41598-024-51344-9 · Cited by 30
Communication-Efficient Online Federated Learning Strategies for Kernel Regression
Gogineni, V. C., Werner, S., Huang, Y.-F., & Kuh, A. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3218484 · Cited by 26
“Furthermore, the proposed partial-sharing-based FL strategies demonstrate good resilience to model-poisoning attacks without additional robust aggregation or adversarial detection mechanisms, as in [36] and [37].”
BackgroundPrivacy-Preserving and Poisoning-Defending Federated Learning in Fog Computing
Li, Y., Zhang, S., Chang, Y. Z., Xu, G., & Li, H. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3302795 · Cited by 23
“In this article, for guaranteeing the model convergence and supporting computation for negative integers, we utilize the method [27] to achieve the quantization.”
MethodByzantines Can Also Learn From History: Fall of Centered Clipping in Federated Learning
Özfatura, K., Özfatura, E., Küpçü, A., & Gündüz, D. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3345171 · Cited by 26
“Hence, ultimately, the target is to make FL secure, private [10], [47], [57], and robust despite data heterogeneity [12], [38].”
BackgroundFederated Learning Methods for Analytics of Big and Sensitive Distributed Data and Survey
Staňo, M., Hluchý, L., Bobák, M., Krammer, P. H., & Tran, V. (2023) · IEEE International Symposium on Applied Computational Intelligence and Informatics (SACI) · doi:10.1109/saci58269.2023.10158622 · Cited by 20
“This makes Federated Learning an attractive approach for applications that require privacy and security [11].”
Metadata
- Authors
- So, J., Guler, B., & Avestimehr, A. S.
- Affiliations
- University of Southern California (US)
- Venue
- IEEE Journal on Selected Areas in Communications
- Published
- 2021-07
- Volume
- 39
- Issue
- 7
- Pages
- 2168-2181
- ISSN
- 0733-8716 (ISSN-L)
- 1558-0008
- OpenAlex
- W3107632872
- Cited by
- 313
- References
- 48
- Citations by year
- 2026: 32 · 2025: 81 · 2024: 67 · 2023: 55 · 2022: 43 · 2021: 27
- FWCI
- 13.06
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Journal on Selected Areas in Communications (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Defense Advanced Research Projects Agency (HR001117C0053)
- Army Research Office (W911NF1810400)
- NSF (CCF-1703575, CCF-1763673)
- Office of Naval Research (N00014-16-1-2189)
- research gifts from Intel and Facebook
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
- Physical Sciences › Computer Science › Information Systems › Cloud Data Security Solutions
- Keywords
- secure federated learning, Byzantine resilience, secure aggregation, stochastic quantization, privacy preservation, convergence guarantees, user dropout
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.