Federated Learning for Privacy Preserving Fraud Detection across Financial Institutions: Architecture Protocols and Operational Governance
DOI:
https://doi.org/10.63282/3050-922X.IJERET-V5I2P111Keywords:
Federated Learning, Fraud Detection, Privacy Secure Aggregation, Differential Privacy, Financial Crime, Anti Money Laundering, Credit Card Fraud GovernanceAbstract
Fraud detection is a shared problem across banks payment processors and fintech partners yet high quality models often require cross institution signals that cannot be centralized due to privacy regulation contractual constraints and competitive boundaries. Federated learning enables collaborative model training without moving raw data by sending model updates rather than records. However practical deployment in financial networks requires more than basic model averaging. It requires robust privacy protection against inference, strong security controls against malicious participants, a design that addresses non independent data distributions and a governance model that aligns with risk and compliance. This paper proposes an architecture led framework for privacy preserving federated fraud detection across financial institutions. We describe system components for horizontal and vertical federation, secure aggregation to protect individual client updates and differential privacy to bound leakage. We present model design options including tabular models representation learning and federated graph learning for transaction relationship signals. We then define an operational lifecycle that covers onboarding, auditability, drift detection, incident response and model rollback. Finally we outline an evaluation plan that measures detection lift, fairness, privacy, resilience and cost. The goal is a blueprint that allows banks and fintech partners to share learning value while preserving customer privacy and supporting regulatory defensibility
References
[1] P. Kairouz et al., "Advances and Open Problems in Federated Learning," Foundations and Trends in Machine Learning, vol. 14, no. 1 2, pp. 1 210, 2021, doi: 10.1561/2200000083.
[2] K. Bonawitz et al., "Practical Secure Aggregation for Privacy Preserving Machine Learning," in Proceedings of the ACM SIGSAC Conference on Computer and Communications Security, 2017, doi: 10.1145/3133956.3133982.
[3] Wang, L., Jia, R., & Song, D. (2020). D2P‑Fed: Differentially private federated learning with efficient communication. arXiv. https://doi.org/10.48550/arXiv.2006.13039
[4] Gunda, S. K. G. (2023). The Future of Software Development and the Expanding Role of ML Models. International Journal of Emerging Research in Engineering and Technology, 4(2), 126-129. https://doi.org/10.63282/3050-922X.IJERET-V4I2P113
[5] W. Yang et al., "FFD: A Federated Learning Based Method for Credit Card Fraud Detection," in Advances in Knowledge Discovery and Data Mining, 2019, doi: 10.1007/978-3-030-23551-2_2.
[6] Zheng, W., Yan, L., Gou, C., & Wang, F.‑Y. (2020). Federated meta‑learning for fraudulent credit card detection. In Proceedings of the Twenty‑Ninth International Joint Conference on Artificial Intelligence (IJCAI) (pp. 4654–4660). https://doi.org/10.24963/ijcai.2020/642
[7] Pan, Z., Wang, G., Li, Z., Chen, L., Bian, Y., & Lai, Z. (2023). 2SFGL: A simple and robust protocol for federated graph‑based fraud detection. arXiv. https://doi.org/10.48550/arXiv.2310.08335
[8] N. F. Aurna et al., "Federated Learning Based Credit Card Fraud Detection: Performance Analysis with Sampling Methods and Deep Learning Algorithms," in 2023 IEEE International Conference on Cyber Security and Resilience, 2023, doi: 10.1109/CSR57506.2023.10224978.
[9] Gunda SK, Yettapu SDR, Bodakunti S, Bikki SB. Decision Intelligence Methodology for AI-Driven Agile Software Lifecycle Governance and Architecture-Centered Project Management, 2023 Mar. 30;4(1):102-8. https://doi.org/10.63282/3050-9262.IJAIDSML-V4I1P112
[10] H. Brendan McMahan et al., "Learning Differentially Private Recurrent Language Models," in International Conference on Learning Representations, 2018, doi: 10.48550/arXiv.1710.06963.
[11] R. Shokri, M. Stronati, C. Song and V. Shmatikov, "Membership Inference Attacks Against Machine Learning Models," in 2017 IEEE Symposium on Security and Privacy, 2017, doi: 10.1109/SP.2017.41.
[12] L. Song, R. Shokri and P. Mittal, "Privacy Risks of Securing Machine Learning Models Against Adversarial Examples," in Proceedings of ACM CCS, 2017, doi: 10.1145/3133956.3134066.
[13] J. Konečný, H. B. McMahan, D. Ramage and P. Richtárik, "Federated Optimization: Distributed Machine Learning for On Device Intelligence," 2016, doi: 10.48550/arXiv.1610.02527.
[14] Alsaeedi, A., & Khan, M. (2019). Software defect prediction using supervised machine learning and ensemble techniques: A comparative study. Journal of Software Engineering and Applications, 12, 85–100. https://doi.org/10.4236/jsea.2019.125007
[15] M. Jagielski et al., "Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning," in 2018 IEEE Symposium on Security and Privacy, 2018, doi: 10.1109/SP.2018.00021.
[16] P. Blanchard, E. M. El Mhamdi, R. Guerraoui and J. Stainer, "Machine Learning with Adversaries: Byzantine Tolerant Gradient Descent," in NeurIPS, 2017, doi: 10.48550/arXiv.1703.02757.
[17] E. Bagdasaryan, A. Veit, Y. Hua, D. Estrin and V. Shmatikov, "How To Backdoor Federated Learning," in Proceedings of AISTATS, 2020, doi: 10.48550/arXiv.1807.00459.
[18] Shah, H., & Sahatiya, P. (2022). A comparative analysis study of software defect prediction using machine learning algorithms on NASA datasets. Journal of Harbin Institute of Technology, 54(9), 67–76. Retrieved from http://hebgydxxb.periodicales.com/index.php/JHIT/article/view/1301
[19] S. Ramaswamy, R. Mathews, K. Rao, and F. Beaufays, "Federated Learning for Emoji Prediction in a Mobile Keyboard," arXiv preprint, 2019, doi: 10.48550/arXiv.1906.04329.
[20] H. Li, K. Ota, and M. Dong, "Learning IoT in Edge: Deep Learning for the Internet of Things with Edge Computing," IEEE Network, vol. 32, no. 1, pp. 96–101, 2018, doi: 10.1109/MNET.2018.1700202.
[21] Al‑Nawaiseh, O., & Hammad, M. (2022). Software defect prediction using stacking generalization of optimized tree‑based ensembles. Applied Sciences, 12(9), 4577. https://doi.org/10.3390/app12094577
[22] Z. Ma, H. Hu, Y. Li, and J. Liu, "Privacy-Enhancing Technologies in Federated Learning: A Survey," Information Fusion, vol. 92, 2023, doi: 10.1016/j.inffus.2022.08.016.
[23] L. Xu, W. Wang, and Z. Qin, "FedSafe: A Federated Learning Framework for Secure and Fair Model Training," IEEE Transactions on Information Forensics and Security, vol. 18, pp. 4123–4135, 2023, doi: 10.1109/TIFS.2023.3284442.
[24] A. G. Howard et al., "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications," arXiv preprint, 2017, doi: 10.48550/arXiv.1704.04861.
[25] S. Wang, T. Tuor, T. Salonidis, K. Leung, C. Makaya, T. He, and K. Chan, "Adaptive Federated Learning in Resource Constrained Edge Computing Systems," IEEE Journal on Selected Areas in Communications, vol. 37, no. 6, pp. 1205–1221, 2019, doi: 10.1109/JSAC.2019.2904348.
[26] M. Chen, Z. Yang, W. Saad, C. Yin, and M. Shikh-Bahaei, "A Joint Learning and Communications Framework for Federated Learning over Wireless Networks," IEEE Transactions on Wireless Communications, vol. 20, no. 1, pp. 269–283, 2021, doi: 10.1109/TWC.2020.3024620.
[27] Shah, H., & Sahatiya, P. (2022). A comparative analysis study of software defect prediction using machine learning algorithms on NASA dataset. Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology, 54(9), 67–76. Retrieved from http://hebgydxxb.periodicales.com/index.php/JHIT/article/view/1301
