AccScience Publishing / IJOCTA / Online First / DOI: 10.36922/IJOCTA025480219
Submit to IJOCTA
Cite this article
1
Download
20
Views
Journal Browser
Volume | Year
Issue
Search
Forthcoming Issue
Current Issue
View All
News and Announcements
View All
RESEARCH ARTICLE

TrustFCL: Task-grained model assessment and aggregation for secure federated continual learning

Xiaoning Wu1 ,  Siqi Du1 ,  Ke Qiu1 ,  Shilin Wen2 ,  Haiting Hou1 ,  Yuxiao Liu1 ,  Jianxin Zhao1 ,  Chi Harold Liu1 ,  Rui Han1*
Show Less
1 Department of Computer Science, Beijing Institute of Technology, Beijing, China
2 North Automatic Control Technology Institute, Taiyuan, Shanxi, China
IJOCTA, 025480219 https://doi.org/10.36922/IJOCTA025480219
Received: 30 November 2025 | Revised: 13 February 2026 | Accepted: 14 February 2026 | Published online: 28 April 2026
© 2026 by the Author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Download PDF
XML
Cite
Abstract

Federated learning (FL) enables decentralized model training across devices, while federated continual learning (FCL) continuously adapts to evolving tasks (varying in categories, data distributions, or problem domains). Currently, FL security issues, such as model poisoning and privacy leaks, are undermining trust and reliability, prompting the development of defense methods. However, with evolving tasks, when a poisoning attack occurs, traditional FL defense methods only evaluate whether the latest model from each client is poisoned. Therefore, existing approaches face two limitations: (i) defense methods based on static tasks ensure security by restricting parameter mutations, which undermines the model’s ability to adapt to new tasks and (ii) knowledge preservation mechanisms (e.g., parameters or gradients), crucial for alleviating forgetting, also introduce new vulnerabilities. Poisoned historical knowledge could be reused to launch attacks that are persistent and difficult to trace. To enable secure and adaptive FCL, this paper proposes TrustFCL, which proposes inter-task reliability and task knowledge reliability as both optimization constraints and control feedback. These reliabilities guide both local anti-forgetting training and global decentralized secure aggregation through a dynamic review committee. Furthermore, TrustFCL stores task reliability information on blockchain to prevent tampering and utilizes the consensus mechanism for privacy-preserving model aggregation. Evaluation results show that TrustFCL reduces the accuracy degradation by 37.1% compared to existing defense methods, and achieves a 17.8% improvement over FCL baselines.

Keywords
Federated continual learning
Task-grained aggregation
Distributed optimization
Multi-agent control
Security
Funding
This paper was supported by National Natural Science Foundation of China (Grant No. 62272046, 62132019, 62302039,61872337), and the Special Program for High-Quality Development of the Ministry of Industry and Information Technology (No. CEIEC-20240), a Cooperative Project with the Northern Institute of Automatic Control Technology.
Conflict of interest
The authors declare they have no competing interests.
References
  1. McMahan B, Moore E, Ramage D, Hampson S, y Arcas BA. Communication-efficient learning of deep networks from decentralized data. Paper presented  at:  proceedings  of  the 20th  International  Conference  on Artificial Intelligence and Statistics; April 20-22, 2017; Fort Lauderdale, FL, USA. Accessed March 17, 2026. https://proceedings.mlr.press/v54/mcmahan17a.html

 

  1. Usynin D, Ziller A, Makowski M, et al. Adversarial interference and its mitigations in privacy-preserving  collaborative  machine learning. Nat Mach Intell. 2021;3(9):749-758. https://www.doi.org/10.1038/s42256-021-00390-3

 

  1. Zhang J, Chen B, Cheng X, Binh HTT., Yu   PoisonGAN:  Generative  poisoning attacks against federated learning in edge computing systems. IEEE Internet Things J. 2020;8(5):3310-3322. https://www.doi.org/10.1109/jiot.2020.3023126

 

  1. Fang M, Cao X, Jia J, Gong NZ. Local model poisoning attacks to Byzantine-robust federated learning. In: Proc of the 29th USENIX Security Symposium (USENIX Security 20); August 12-14, 2020;  USENIX Association; 2020:1605-1622. Available from https://www.usenix.org/system/files/sec20-fang.pdf

 

  1. Ren H, Deng J, Xie X. Grnn: generative regression neural network—a data leakage attack for federated learning. ACM Trans Intell Syst Technol. 2022;13(4):1-24. https://www.doi.org/10.1145/3510032

 

  1. Robins A. Catastrophic forgetting, rehearsal and pseudorehearsal. Connect Sci. 1995;7(2):123-146. https://www.doi.org/10.1080/09540099550039318

 

  1. Wang H, Sreenivasan K, Rajput S, et al. Attack of the Tails: Yes, You Really Can Backdoor Federated Learning. arXiv. Preprint online July 9, 2020. https://www.doi.org/10.48550/ARXIV.2007.05084

 

  1. Cao X, Fang M, Liu J, Gong NZ. Fltrust: Byzantine-robust federated learning via trust bootstrapping. Paper presented at: The Network and Distributed System Security Symposium 2021; February 21-25, 2021; Virtual. Accessed March 17, 2026. https://www.ndss-symposium.org/ndss-paper/fltrust-byzantine-robust-federated-learning-via-trust-bootstrapping/

 

  1. Mao Y, Yuan X, Zhao X, Zhong S. Romoa: Robust model aggregation for the resistance of federated learning to model poisoning attacks. Proc Eur Symp Res Comput Secur. 2021;476-496. https://www.doi.org/10.1007/978-3-030-88418-5 23

 

  1. Geng G, Cai T, Yang Z. Better safe than sorry: Constructing byzantine-robust federated learning with synthesized trust. 2023;12(13):2926. https://www.doi.org/10.3390/electronics12132926

 

  1. Chen Y, Su L, Xu J. Distributed statistical machine learning in  adversarial settings: Byzantine gradient descent. Proc ACM Meas Anal Comput Syst. 2017;1(2):1-25. https://www.doi.org/10.1145/3154503

 

  1. Blanchard P, El Mhamdi EM, Guerraoui R, Stainer J. Machine learning with adversaries: Byzantine tolerant gradient descent. Adv Neural Inf Process Syst. 2017;30:118-128. https://www.doi.org/10.5555/3294771.3294783

 

  1. Nikishin E, Oh  J,  Ostrovski  G,  et  Deep  reinforcement  learning  with  plasticity injection.  Adv  Neural  Inf  Process  Syst. 2023;36:37142-37159. https://www.doi.org/10.5555/3666122.3667736

 

  1. Dong J, Wang L, Fang Z, et al. Federated class-incremental learning. arXiv. Preprint online March 22, 2022. https://www.doi.org/10.48550/arXiv.2203.11473

 

  1. Hegedu˝s I, Danner G, Jelasity M. Gossip learning as a decentralized alternative to federated learning. Proc IFIP Int Conf Distrib Appl Interoperable Syst. 2019;74-90. https://www.doi.org/10.1007/978-3-030-22496-7 5

 

  1. Hamedi P, Razavi-Far R, Hallaji E. Federated continual learning: concepts, challenges, and solutions. 2025;651(C). https://www.doi.org/10.1016/j.neucom.2025.130844

 

  1. Birashk A, Khan L. Federated continual learning for task-incremental and  class-incremental problems:  A survey. Expert Syst Appl. 2025;129278.  https://www.doi.org/10.1016/j.eswa.2025.129278

 

  1. Yoon J, Jeong W, Lee G, Yang E, Hwang SJ. Federated continual learning with weighted inter-client Paper  presented  at: proceedings of the 38th International Conference on Machine Learning; July 18-24, 2021; Virtual. Accessed March 17, 2026.  https://proceedings.mlr.press/v139/yoon21b.html

 

  1. Li J, Shao Y, Wei K, et al. Blockchain assisted decentralized federated learning (BLADE-FL): Performance analysis and resource allocation. IEEE Trans. Parallel Distrib Syst. 2021;33(10):2401-2415. https://www.doi.org/10.1109/tpds.2021.3138848

 

  1. Li J, Shao Y, Wei K, et al. Blockchain Assisted Decentralized Federated  Learning (BLADE-FL):  Performance  Analysis and Resource Allocation. IEEE Trans Parallel Distrib Syst. 2022;33(10):2401-2415. https://www.doi.org/10.1109/tpds.2021.3138848

 

  1. Singh S, Rathore S, Alfarraj O, Tolba A, Yoon B. A framework for privacy-preservation of IoT healthcare data using Federated Learning and blockchain technology. Future Gener Comput Syst. 2022;129:380-388. https://www.doi.org/10.1016/j.future.2021.11.028

 

  1. Lu Y, Huang X, Zhang K, Maharjan S, Zhang Y. Blockchain empowered asynchronous federated learning for secure data sharing in internet of vehicles. IEEE Trans Veh Technol. 2020;69(4):4298-4311. https://www.doi.org/10.1109/TVT.2020.2973651

 

  1. Ma C, Li J, Shi L, et al. When federated learning meets blockchain: A new distributed learning paradigm. IEEE Comput Intell Mag. 2022;17(3):26-33. https://www.doi.org/10.1109/mci.2022.3180932

 

  1. Qi J, Lin F, Chen Z, Tang C, Jia R, Li M. High-quality model  aggregation for blockchain-based  federated  learning via reputation-motivated task participation. IEEE Internet Things J. 2022;9(19):18378-18391. https://www.doi.org/10.1109/jiot.2022.3160425

 

  1. Gao L, Li L, Chen Y, Xu CZ, Xu M. FGFL: a blockchain-based fair incentive governor for Federated Learning. Parallel Distrib Comput. 2022;163:283-299. https://www.doi.org/10.1016/j.jpdc.2022.01.019

 

  1. Guo S, Zhang K, Gong B, et al. Sandbox computing: A data privacy trusted sharing paradigm via blockchain and federated learning. IEEE Trans Comput. 2022;72(3):800-810. https://www.doi.org/10.1109/tc.2022.3180968

 

  1. Li Z, Yu H, Zhou T, et al. Byzantine resistant secure blockchained federated learning at the edge. IEEE Netw. 2021;35(4):295-301. https://www.doi.org/10.1109/mnet.011.2000604

 

  1. Abdel-Basset M, Moustafa  N,  Hawash  Privacy-preserved   cyberattack   detection in  Industrial  Edge  of  Things  (IEoT): A blockchain-orchestrated   federated   learning approach.   IEEE   Trans   Ind   Inform. 2022;18(11):7920-7934. https://www.doi.org/10.1109/tii.2022.3167663

 

  1. Li T, Sahu  AK,  Zaheer  M,  Sanjabi  M, Talwalkar A, Smith V. Federated optimization in heterogeneous networks. Proc Mach Learn Syst. 2020;2:429-450. https://proceedings.mlsys.org/paper/2020/hash/1f5fe83998a09396ebe6477d9475ba0c-Abstract.html

 

  1. Roberts L, Smyth, E. A simplified convergence theory for Byzantine resilient stochastic gradient descent. EURO J Comput Optim. 2022;10:100038. https://www.doi.org/10.1016/j.ejco.2022.100038

 

  1. Chen B, Fan L, Zeng X, Gu M, Zhou, J. A contribution-aware federated framework for electric vehicle batteries health estimation. IEEE Internet Things J. https://www.doi.org/10.1109/jiot.2024.3524005

 

  1. Gupta A, Luo T, Ngo MV, Das SK. Long-Short History of Gradients Is All You Need: Detecting Malicious and Unreliable Clients in Federated Learning. In: Lecture Notes in Computer Science. Springer Nature Switzerland. 2022:445-46. https://www.doi.org/10.1007/978-3-031-17143-7 22

 

  1. Chen C, Li  P,  Sakurai  K,  et  A multi-head   federated   continual   learning approach   for   improved   flexibility   and robustness   in   edge   environments. Int J   Netw   Comput. 2024;14(2):123-144. https://www.doi.org/10.15803/ijnc.14.2 123

 

  1. Chen L, Zhao D, Tao L, et al. A credible and fair federated learning framework based on IEEE  Trans  Artif  Intell. 2024;6(2):301-316. https://www.doi.org/10.1109/tai.2024.3355362

 

  1. Kang J, Xiong Z, Niyato D, Xie S, ZhangJ. Incentive mechanism for reliable federated learning: A joint optimization approach to combining reputation and contract theory. IEEE Internet Things J. 2019;6(6):10700-10714. https://www.doi.org/10.1109/jiot.2019.2940820

 

  1. Zouari F, Mahmud M. Neural Network-Based Robust Adaptive Output Feedback Control for MIMO Time-Varying  Delay  Paper presented at:  the Global Conference on Applications of Artificial Intelligence and Data Science 2024; July 22-26, 2024; Munich, Germany, and virtually. Accessed March 17, 2026. https://www.doi.org/10.1007/978-3-031-98498-3 5

 

  1. Boulkroune A, Boubellouta A, Bouzeriba A, Zouari F. Practical finite-time fuzzy synchronization of chaotic  systems with non-integer  orders:  Two  chattering-free approaches.  J  Syst  Sci  Syst Eng. 2025;34(3):334-359. https://www.doi.org/10.1007/s11518-024-5635-7

 

  1. Jiang Y, Shen J, Liu Z, Tan CW, Lam KY. Towards efficient and certified recovery from poisoning attacks in federated learning. IEEE Trans Inf Forensics Secur. https://www.doi.org/10.1109/tifs.2025.3533907

 

  1. Krizhevsky A, Hinton G. Learning multiple layers of features from tiny images. Univ Toronto; 2009. Available from https://www.cs.toronto.edu/ kriz/learning-features-2009-TR.pdf

 

  1. Vinyals O, Blundell C, Lillicrap T, et al. Matching networks for one shot learning. Adv Neural Inf Process Syst. 2016;29. https://www.doi.org/10.5555/3157382.3157504

 

  1. Le Y, Yang    Tiny  imagenet visual  recognition  challenge. Stanford Univ  CS231N ;  2015.  Available  from https://cs231n.stanford.edu/reports/2015/pdfs/yleproject.pdf

 

  1. Ke Z, Liu B, Wang H, Shu L. Continual learning with knowledge transfer for sentiment classification. Proc Jt Eur Conf Mach Learn Knowl Discov   2020;683-698. https://www.doi.org/10.1007/978-3-030-67664-3 41

 

  1. Collins L, Hassani H, Mokhtari A, Shakkottai S. Exploiting shared representations for personalized  federated    Paper presented at:  proceedings of the 38th International  Conference  on  Machine Learning. 2021;18-24; Virtual. Available from https://proceedings.mlr.press/v139/collins21a.html

 

  1. Deng Y, Kamani  MM,  Mahdavi M.  Adaptive  personalized federated learning. arXiv. Preprint. Available from https://www.doi.org/10.48550/arXiv.2003.13461

 

  1. He C, Annavaram M, Avestimehr S. Group knowledge transfer: Federated learning of large cnns at the edge. Adv Neural Inf Process Syst. 2020;33:14068-14080. https://www.doi.org/10.5555/3495724.3496904

 

  1. Luopan Y, Han R, Zhang Q, Liu CH, Wang G, Chen LY. Fedknow: Federated continual learning with signature task knowledge integration at edge. Proc IEEE 39th Int Conf Data Eng. 2023;341-354. https://www.doi.org/10.1109/icde55515.2023.00033

 

  1. Zuo X, Luopan Y, Han R, et al. FedViT: federated continual learning of vision transformer at edge. Future Gener Comput Syst. 2024;154:1-15. https://www.doi.org/10.1016/j.future.2023.11.038

 

  1. Wang Q, Liu B, Li Y. Traceable federated continual learning. Proc IEEE/CVF Conf Comput Vis Pattern Recognit. 2024;12872-12881. https://www.doi.org/10.1109/cvpr52733.2024.01223

 

  1. Wuerkaixi A, Cui S, Zhang J, et al. Accurate Forgetting for Heterogeneous Federated Continual Learning. arXiv. Preprint online Feb 20, 2025. https://www.doi.org/10.48550/arXiv.2502.14205

 

  1. Sattler F, Mu¨ller KR, Samek W. Clustered federated learning: model-agnostic distributed multitask optimization under privacy constraints. IEEE Trans Neural  Netw  Learn  2020;32(8):3710-3722. https://www.doi.org/10.1109/tnnls.2020.3015958

 

  1. Ghosh A, Chung J, Yin D, Ramchandran K. An Efficient Framework for Clustered Federated Learning. IEEE Trans Inform Theory. 2022;68(12):8076-8091. https://www.doi.org/10.1109/tit.2022.3192506

 

  1. Tong G, Li GB, Wu J, Li J. Gradmfl: gradient memory-based federated learning for hierarchical knowledge transferring over non-iid data. Proc Int Conf Algorithms Archit Parallel Process. 2021;612-626. https://www.doi.org/10.1007/978-3-030-95384-3 38

 

  1. Sun Z, Kairouz P, Suresh AT, McMahan HB. Can you really backdoor federated learning? arXiv. Preprint online November 18, 2019. https://www.doi.org/10.48550/arXiv.1911.07963

 

  1. Sun J, Li A, DiValentin L, Hassanzadeh A, Chen Y, Li H. Fl-wbc: enhancing robustness against model poisoning attacks in federated learning from a client perspective. Adv Neural Inf Process Syst. 2021;34:12613-12624. https://www.doi.org/10.5555/3540261.3541226

 

  1. Xie C, Koyejo O, Gupta I. SLSGD: secure and efficient distributed on-device machine learning. Paper presented at: joint European Conference on Machine Learning and Knowledge Discovery in Databases 2019; September 16-20, 2019; Wu¨rzburg, Germany. Accessed March 17, 2026. https://link.springer.com/chapter/10.1007/978-3-030-46147-8 13

 

  1. Yin D, Chen Y, Kannan R, Bartlett P. Byzantine-robust distributed learning: towards optimal statistical rates. Paper presented at: Proceedings of the 35th International Conference on Machine Learning; July 10-15, 2018; Stockholmsm¨assan, Stockholm, Sweden. Accessed March 17, 2026. https://proceedings.mlr.press/v80/yin18a

 

  1. Karimireddy SP, He L, Jaggi M. Byzantine-Robust Learning on Heterogeneous Datasets via Bucketing. Paper presented at: International Conference on Learning Representations; April 25-29, 2022; Virtually. Accessed March 17, 2026. https://openreview.net/forum?id=jXKKDEi5vJt

 

  1. Han S, Buyukates B, Hu Z, et al. Fedsecurity: a benchmark for  attacks  and  defenses  in federated learning and federated llms. Paper presented at:  Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and  Data  Mining;  August  25-29, 2024; Barcelona, Spain. Accessed March 17, 2026. https://www.doi.org/10.1145/3637528.3671545
Next article in this issue
Share
Back to top
An International Journal of Optimization and Control: Theories & Applications, Electronic ISSN: 2146-5703 Print ISSN: 2146-0957, Published by AccScience Publishing
We use cookies on our website to ensure you get the best experience.
Read more about our cookies here
Accept