AXBOROT TIZIMLARIDA MA’LUMOTLAR MAXFIYLIGINI TA'MINLASHDA FEDERATIV OʻQITISH (FEDERATED LEARNING) YONDASHUVINING SAMARADORLIGI NAZARIY TAHLIL VA QIYOSIY BAHOLASH

G‘ofurova M. G.


Annotatsiya

Zamonaviy axborot tizimlarida ma’lumotlar hajmining jadal oʻsishi bilan birga shaxsiy ma’lumotlarni himoya qilish muammosi tobora dolzarb tus olmoqda. Markazlashtirilgan mashinaviy oʻqitish modellari xom ma’lumotlarni yagona serverga toʻplashni talab qilib, GDPR kabi xalqaro me’yoriy talablar bilan ziddiyat keltirib chiqaradi. Federativ oʻqitish (Federated Learning, FL) — ma’lumotlarni mahalliy qurilmalarda saqlab, faqat model parametrlarini almashinish orqali umumiy model shakllantiruvchi taqsimlangan oʻqitish paradigmasi — ushbu muammoning zamonaviy va ilmiy asoslangan yechimi sifatida qaralmoqda. Ushbu maqolada FL ning nazariy asoslari, asosiy algoritmik mexanizmlari (FedAvg, FedProx, Differensial maxfiylik) va amaliy tatbiq natijalari tizimli adabiyotlar sharhi hamda qiyosiy tahlil metodlari yordamida IMRAD uslubida oʻrganilgan. Tadqiqot natijalari shuni koʻrsatadiki, FL yondashuvi model aniqligini 92-96% darajada saqlab, GDPR talablariga toʻliq muvofiqlikni ta’minlaydi. FedProx algoritmi non-IID muhitlarda FedAvg ga nisbatan 22% gacha ustunlikni namoyish etadi. Kommunikatsiya xarajatlari, non-IID muammo va xavfsizlik tahdidlari hali hal etilmagan cheklovlar boʻlib qolmoqda.


Adabiyotlar

  1. Statista Research Department. Volume of data/information created worldwide from 2010 to 2025. — Hamburg: Statista GmbH, 2023.
  2. European Union. Regulation (EU) 2016/679 (General Data Protection Regulation). — Official Journal of the European Union, L 119, 2018. — P. 1-88.
  3. Kiteworks. GDPR Fines Hit 7.1 Billion EUR: Data Privacy Enforcement Trends in 2026. [Online]. URL: https://www.kiteworks.com (murojaat: 2026-yil mart).
  4. Duality Technologies. Federated Learning and Global Data Sovereignty Compliance. [Online]. URL: https://dualitytech.com/blog/federated-learning (murojaat: 2025).
  5. McMahan H.B., Moore E., Ramage D., Hampson S., y Arcas B.A. Communication-efficient learning of deep networks from decentralized data // AISTATS. — PMLR 54, 2017. — P. 1273-1282.
  6. Li T., Sahu A.K., Talwalkar A., Smith V. Federated learning: Challenges, methods, and future directions // IEEE Signal Processing Magazine. — 2020. — Vol. 37, No. 3. — P. 50-60.
  7. Oʻzbekiston Respublikasi. «Shaxsga doir ma’lumotlar toʻgʻrisida»gi Qonun, OʻRQ-547-son. — Toshkent, 2019.
  8. Li T., Sahu A.K., Zaheer M. et al. Federated optimization in heterogeneous networks (FedProx) // MLSys. — 2020. — Vol. 2. — P. 429-450.
  9. Abadi M., Chu A., Goodfellow I. et al. Deep learning with differential privacy // CCS'16. — ACM, 2016. — P. 308-318.
  10. Gosselin R., Vieu L., Loukil F., Benoit A. The federation strikes back: A survey of federated learning privacy attacks, defenses, applications // ACM Computing Surveys. — 2025. DOI: 10.1145/3724113.
  11. Roth H.R., Chang K., Singh P. et al. Federated learning for breast density classification: A real-world implementation // DART, LNCS. — Vol. 12444. — Springer, 2020. — P. 181-191.
  12. Teo Z.L., Jin L., Liu N. et al. Federated machine learning in healthcare: A systematic review // Cell Reports Medicine. — 2024. — Vol. 5, No. 2. DOI: 10.1016/j.xcrm.2024.101419.
  13. Hagestedt I. et al. Toward a tipping point in federated learning in healthcare and life sciences // Patterns. — 2024. — Vol. 5, No. 11. — P. 101077.
  14. Collins E., Wang M. Federated learning: A survey on privacy-preserving collaborative intelligence. arXiv:2504.17703. — Arizona State University, 2025.
  15. Page M.J. et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews // BMJ. — 2021. — Vol. 372. — n71.
  16. Karimireddy S.P. et al. SCAFFOLD: Stochastic controlled averaging for federated learning // ICML. — PMLR 119, 2020. — P. 5088-5099.
  17. McMahan H.B., Ramage D. Federated learning: Collaborative machine learning without centralized training data // Google AI Blog. — 2017.