Proceedings of the International scientific and practical conference ―Science, Technology and Culture: Dynamics of Change in the XXI Century‖ (December 1921, 2025) / Publisher website: www.naukainfo.com. – Baku, Azerbaijan, 2026. – 90 p.

24 value of the study lies in the developed methodology, which can be directly applied in critical domains where data protection is a priority, such as mobile healthcare applications and financial information systems. Finally, directions for future research are outlined, including the reduction of computational overhead, the development of methods for dynamically adjusting the privacy parameter $\epsilon$ during training, and the integration of the proposed mechanism with other protection techniques, such as homomorphic encryption or trusted execution environments (TEE). REFERENCES: 1. McMahan B., Moore E., Ramage D., Hampson S., Arcas B. A. y. Communication-Efficient Learning of Deep Networks from Decentralized Data // Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS). – 2017. – P. 1273–1282. – URL: https://arxiv.org/abs/1602.05629 2. Dwork C., Roth A. The Algorithmic Foundations of Differential Privacy. – Hanover : Now Publishers Inc., 2014. – 261 p. – URL: https://www.cis.upenn.edu/~aaroth/Papers/privacybook.pdf 3. Abadi M., Chu A., Goodfellow I. et al. Deep Learning with Differential Privacy // Proceedings of the ACM SIGSAC Conference on Computer and Communications Security (CCS). – 2016. – P. 308–318. – URL: https://arxiv.org/abs/1607.00133 4. Nasr M., Shokri R., Houmansadr A. Comprehensive Privacy Analysis of Deep Learning: Stand-alone and Federated Learning under Passive and Active White- box Inference Attacks // IEEE Symposium on Security and Privacy. – 2019. – P. 739–753. – URL: https://arxiv.org/abs/1812.00910 5. Geyer R. C., Klein T., Nabi M. Differentially Private Federated Learning: A Client Level Perspective // NeurIPS Workshop on Private Multi-Party Machine Learning. – 2017. – URL: https://arxiv.org/abs/1712.07557