Adaptive Trust-Aware Post-Quantum Secure Federated Learning with Dynamic Differential Privacy

Year : 2026 | Volume : 13 | Issue : 02 | Page : 38 44
By

Nikita S. Shinde,

Nita Thakare,

  1. Student, Department of Computer Engineering (AIDS), Guru Gobind Singh College of Engineering and Research Centre, Jawahar Education Society’s, JIT, Maharashtra, India
  2. Professor, Department of Computer Engineering (AIDS), Guru Gobind Singh College of Engineering and Research Centre, Jawahar Education Society’s, JIT, Maharashtra, India

Abstract

FL allows multiple users to train the machine learning model and ensure that they cannot share their raw data. Post-quantum secure aggregation and differential privacy for FL are introduced by the Beskar Framework; it assumes that all selected nodes will perform reliably. Due to hardware faults, network failures, or malicious intent, some of the assisting nodes may perform unreliably while we are developing it in the real-world. To assist the nodes, this paper proposes a framework known as Beskar-AT (Adaptive Trust), which provides a trust scoring mechanism. The model observes the recent activities of each node and then assigns a trust value to each of the assisting nodes. Low-score nodes are temporarily removed from the aggregation process and replaced with a rotating pool. The proposed model automatically changes the privacy level based on the actual number of trusted nodes participating in the process. These two combined features help to improve the reliability and the privacy-utility balance of post-quantum secure FL. With the experiment of MNIST, CIFAR-10, and CIFAR-100 show that the system named Beskar-AT maintains high model accuracy and lower noise overhead compared to the original Beskar framework, especially when some of the assisting nodes behave irregularly.

Keywords: Federated learning, post-quantum cryptography, dynamic differential privacy, trust-based management, secure model aggregation, adaptive noise mechanism

[This article belongs to Journal of Advanced Database Management & Systems ]

How to cite this article: Nikita S. Shinde, Nita Thakare. Adaptive Trust-Aware Post-Quantum Secure Federated Learning with Dynamic Differential Privacy. Journal of Advanced Database Management & Systems. 2026; 13(02):38-44.
How to cite this URL: Nikita S. Shinde, Nita Thakare. Adaptive Trust-Aware Post-Quantum Secure Federated Learning with Dynamic Differential Privacy. Journal of Advanced Database Management & Systems. 2026; 13(02):38-44. Available from: https://journals.stmjournals.com/joadms/article=2026/view=259061

References

  1. Carlini N, Tramèr F, Wallace E, Jagielski M, Herbert-Voss A, Lee K, et al. Extracting training data from large language models. In: Proceedings of the 30th USENIX Security Symposium (USENIX Security 21); 2021. p. 2633-2650.
  2. Shokri R, Stronati M, Song C, Shmatikov V. Membership inference attacks against machine learning models. In: 2017 IEEE Symposium on Security and Privacy (SP); 2017. p. 3-18. doi:10.1109/SP.2017.41.
  3. Zhang Y, Behnia R, Yavuz AA, Ebrahimi R, Bertino E. Efficient full-stack private federated deep learning with post-quantum security. IEEE Trans Dependable Secure Comput. 2025;22(5):5567-5583. doi:10.1109/TDSC.2025.3568704.
  4. Segal A, Marcedone A, Kreuter B, Ramage D, McMahan HB, Seth K, et al. Practical secure aggregation for privacy-preserving machine learning. In: Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security; 2017. p. 1175-1191. doi:10.1145/3133956.3133982.
  5. Bell JH, Bonawitz KA, Gascón A, Lepoint T, Raykova M. Secure single-server aggregation with (poly)logarithmic overhead. In: Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security; 2020. p. 1253-1269. doi:10.1145/3372297.3417885.
  6. Ma Y, Woods J, Angel S, Polychroniadou A, Rabin T. Flamingo: multi-round single-server secure aggregation with applications to private federated learning. In: 2023 IEEE Symposium on Security and Privacy (SP); 2023. p. 477-496. doi:10.1109/SP46215.2023.10179434.
  7. Behnia R, Riasi A, Ebrahimi R, Chow SSM, Padmanabhan B, Hoang T. Efficient secure aggregation for privacy-preserving federated machine learning. In: 2024 Annual Computer Security Applications Conference (ACSAC); 2024. p. 778-793. doi:10.1109/ACSAC63791.2024.00069.
  8. Yang S, Chen Y, Tu S, Yang Z. A post-quantum secure aggregation for federated learning. In: Proceedings of the 2022 12th International Conference on Communication and Network Security; 2022. p. 117-124. doi:10.1145/3586102.3586120.
  9. Chamikara MAP, Liu D, Camtepe S, Nepal S, Grobler M, Bertok P, et al. Local differential privacy for federated learning [Preprint]. arXiv:2202.06053; 2022. doi:10.48550/arXiv.2202.06053.
  10. Truex S, Liu L, Chow KH, Gursoy ME, Wei W. LDP-Fed: federated learning with local differential privacy. In: Proceedings of the Third ACM International Workshop on Edge Systems, Analytics and Networking; 2020. p. 61-66. doi:10.1145/3378679.3394533.
  11. Kairouz P, McMahan HB, et al. Advances and open problems in federated learning. Found Trends Mach Learn. 2021;14(1-2):1-210. doi:10.1561/2200000083.
  12. Abadi M, Chu A, Goodfellow I, McMahan HB, Mironov I, Talwar K, et al. Deep learning with differential privacy. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security; 2016. p. 308-318. doi:10.1145/2976749.2978318.
  13. Yang Y, Hui B, Yuan H, Gong N, Cao Y. PrivateFL: accurate, differentially private federated learning via personalized data transformation. In: Proceedings of the 32nd USENIX Security Symposium (USENIX Security 23); 2023. p. 1595-1612.
  14. Jøsang A, Ismail R, Boyd C. A survey of trust and reputation systems for online service provision. Decis Support Syst. 2007;43(2):618-644. doi:10.1016/j.dss.2005.05.019.
  15. Mironov I. Rényi differential privacy. In: 2017 IEEE 30th Computer Security Foundations Symposium (CSF); 2017. p. 263-275. doi:10.1109/CSF.2017.11.
  16. Ducas L, Kiltz E, Lepoint T, Lyubashevsky V, Schwabe P, Seiler G, et al. CRYSTALS-Dilithium: a lattice-based digital signature scheme. IACR Trans Cryptogr Hardw Embed Syst. 2018;2018(1):238-268. doi:10.13154/tches.v2018.i1.238-268.
  17. Bos J, Ducas L, Kiltz E, Lepoint T, Lyubashevsky V, Schanck JM, et al. CRYSTALS-Kyber: a CCA-secure module-lattice-based KEM. In: 2018 IEEE European Symposium on Security and Privacy (EuroS&P); 2018. p. 353-367. doi:10.1109/EuroSP.2018.00032.
  18. Beutel DJ, Topal T, Mathur A, Qiu X, Fernandez-Marques J, Gao Y, et al. Flower: a friendly federated learning research framework [Preprint]. arXiv:2007.14390; 2020. doi:10.48550/arXiv.2007.14390.
  19. Dwork C. Differential privacy: a survey of results. In: Agrawal M, Du D, Duan Z, Li A, editors. Theory and Applications of Models of Computation. Berlin: Springer; 2008. p. 1-19. doi:10.1007/978-3-540-79228-4_1.
  20. Di Crescenzo G. Sharing one secret vs. sharing many secrets. Theor Comput Sci. 2003;295(1-3):123-140. doi:10.1016/S0304-3975(02)00399-7.
Regular Issue Subscription Original Research
Volume 13
Issue 02
Received 05/05/2026
Accepted 14/07/2026
Published 22/07/2026
Publication Time 78 Days

Login

My IP

PlumX Metrics

Support