Customer-Level Differential Privacy in Federated Repeat Purchase Prediction with Unequal Retail Clients
Keywords:
Customer behaviour, Differential privacy, Federated learning, Non-IID data, Retail analyticsAbstract
Federated customer prediction requires a clear distinction between decentralised optimisation and a formal privacy mechanism. This study implements customer-level clipped gradient aggregation for repeat-purchase prediction using UCI Online Retail. A fixed 90-day history precedes 1 November 2011, and the outcome records purchases during the following 30 days. A deterministic customer split produces 1,457 training, 497 validation, and 507 test observations. Countries define 27 unequal simulated training clients, with the largest containing 1,316 customers. Each customer contributes one bounded feature vector and one clipped logistic gradient per round. Gaussian noise is added to the aggregate under a trusted coordinator, and a Rényi accountant reports replacement-adjacency privacy bounds for 100 full-participation rounds. Nonprivate federation achieves test ROC AUC of 0.6769 and average precision of 0.6366. At a per-model privacy bound of = 3.194 and = 0.00001, mean ROC AUC is 0.6812 and average precision is 0.6397 across five noise seeds. This small numerical increase is not evidence that privacy improves prediction. The results demonstrate a reproducible privacy-utility calculation for a limited retail cohort, without establishing distributed system scalability or protection against an untrusted aggregation server.
References
D. Chen, “Online Retail,” UCI Machine Learning Repository, 2015.
B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. Agüera y Arcas, “Communication-efficient learning of deep networks from decentralized data,” Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, 2017, pp. 1273–1282.
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, Jun. 2021.
T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, “Federated optimization in heterogeneous networks,” Proceedings of Machine Learning and Systems, 2020, pp. 429–450.
S. P. Karimireddy, S. Kale, M. Mohri, S. Reddi, S. Stich, and A. T. Suresh, “SCAFFOLD: Stochastic controlled averaging for federated learning,” Proceedings of the 37th International Conference on Machine Learning, 2020, pp. 5132–5143.
S. Reddi et al., “Adaptive Federated Optimization,” arXiv, Feb. 2020.
L. Zhu, Z. Liu, and S. Han, “Deep leakage from gradients,” Advances in Neural Information Processing Systems, 2019.
R. Shokri, M. Stronati, C. Song and V. Shmatikov, “Membership inference attacks against machine learning models,” 2017 IEEE Symposium on Security and Privacy (SP), San Jose, CA, USA, 2017, pp. 3–18.
C. Dwork and A. Roth, “The algorithmic foundations of differential privacy,” Foundations and Trends in Theoretical Computer Science, vol. 9, no. 3–4, pp. 211–407, Aug. 2014.
K. Bonawitz et al., “Practical secure aggregation for privacy-preserving machine learning,” Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security, Oct. 2017, pp. 1175–1191.
M. Abadi et al., “Deep learning with differential privacy,” arXiv, Oct. 2016.
I. Mironov, “Rényi differential privacy,” 2017 IEEE 30th Computer Security Foundations Symposium (CSF), Santa Barbara, CA, USA, 2017, pp. 263–275.
Q. Li, Y. Diao, Q. Chen and B. He, “Federated learning on Non-IID Data Silos: An experimental study,” 2022 IEEE 38th International Conference on Data Engineering (ICDE), Kuala Lumpur, Malaysia, 2022, pp. 965–978.
X. Li, K. Huang, W. Yang, S. Wang, and Z. Zhang, “On the convergence of FedAvg on Non-IID Data,” arXiv, Jul. 2019.
J. Wang, Q. Liu, H. Liang, G. Joshi, and H. V. Poor, “Tackling the objective inconsistency problem in heterogeneous federated optimization,” Proceedings of the 34th International Conference on Neural Information Processing Systems, Dec. 2020, pp. 7611–7623.
X. Li, M. Jiang, X. Zhang, M. Kamp, and Q. Dou, “FedBN: Federated learning on Non-IID features via local batch normalization,” arXiv, Feb. 2021.
T. Li, S. Hu, A. Beirami, and V. Smith, “Ditto: Fair and robust federated learning through personalization,” Proceedings of the 38th International Conference on Machine Learning, 2021, pp. 6357–6368.
T. Li, M. Sanjabi, A. Beirami, and V. Smith, “Fair resource allocation in federated learning,” arXiv, May 2019.
T. Gebru et al., “Datasheets for datasets,” Communications of the ACM, vol. 64, no. 12, pp. 86–92, Nov. 2021.
M. Mitchell et al., “Model cards for model reporting,” arXiv, Oct. 2018.
S. Kapoor and A. Narayanan, “Leakage and the reproducibility crisis in machine learning-based science,” Patterns, vol. 4, no. 9, Sept. 2023.
T. Saito and M. Rehmsmeier, “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets,” PLoS ONE, vol. 10, no. 3, 2015.
C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, “On calibration of modern neural networks,” Proceedings of the 34th International Conference on Machine Learning, Aug. 2017, pp. 1321–1330.
P. W. Koh et al., “WILDS: A benchmark of in-the-wild distribution shifts,” Proceedings of the 38th International Conference on Machine Learning, 2021, pp. 5637–5664.
A. N. Angelopoulos and S. Bates, “A gentle introduction to conformal prediction and distribution-free uncertainty quantification,” arXiv, Jul. 2021.
S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” arXiv, Dec. 2017, pp. 4768–4777.
M. T. Ribeiro, S. Singh, and C. Guestrin, ““Why should i trust you?”: Explaining the predictions of any classifier,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug. 2016, pp. 1135–1144.
M. Sundararajan, A. Taly, and Q. Yan, “Axiomatic attribution for deep networks,” Proceedings of the 34th International Conference on Machine Learning, Aug. 2017, pp. 3319–3328.
T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug. 2016, pp. 785–794.
J. Adebayo, J. Gilmer, M. Muelly, I. Goodfellow, M. Hardt, and B. Kim, “Sanity checks for saliency maps,” NIPS'18: Proceedings of the 32nd International Conference on Neural Information Processing Systems, pp. 9525–9536, Dec. 2018.
