Transfer learningning amaliy qo‘llanilishi va cheklovlari
Abstract
Mazkur maqolada sun’iy intellekt va mashinali o‘qitish sohasining zamonaviy yo‘nalishlaridan biri hisoblangan transfer learning texnologiyasining nazariy asoslari, amaliy qo‘llanilishi hamda mavjud cheklovlari tahlil qilinadi. Transfer learning usulining chuqur o‘rganish modellarini qayta tayyorlashdagi samaradorligi, ma’lumotlar yetishmovchiligi sharoitida yuqori aniqlikni ta’minlash imkoniyatlari yoritilgan. Shuningdek, kompyuter ko‘rish, tabiiy tilni qayta ishlash, tibbiyot, telekommunikatsiya va robototexnika sohalarida transfer learning texnologiyasining qo‘llanilish misollari keltirilgan. Modelning moslashuvchanligi, hisoblash resurslariga bo‘lgan talab, domenlar o‘rtasidagi farq hamda “negative transfer” muammolari ilmiy jihatdan tahlil qilingan. Tadqiqot natijalari transfer learning texnologiyasi sun’iy intellekt tizimlarining samaradorligini oshirishda muhim vosita ekanligini ko‘rsatadi.
Kalit so‘zlar: Transfer learning, mashinali o‘qitish, chuqur o‘rganish, sun’iy intellekt, neyron tarmoqlar, kompyuter ko‘rish, tabiiy tilni qayta ishlash, negative transfer, CNN, fine-tuning.
References
1. Pan S. J., Yang Q. A Survey on Transfer Learning // IEEE Transactions on Knowledge and Data Engineering. — 2010. — Vol. 22(10). — P. 1345–1359.
2. Goodfellow I., Bengio Y., Courville A. Deep Learning. — MIT Press, 2016.
3. Brownlee J. Deep Learning for Computer Vision. — Machine Learning Mastery, 2019.
4. Devlin J., Chang M., Lee K., Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding // NAACL, 2019.
5. Ruder S. Neural Transfer Learning for Natural Language Processing. — National University of Ireland, 2019.
6. Chollet F. Deep Learning with Python. — Manning Publications, 2021.
