Russian Federation
The paper examines and compares the main approaches to training machine learning and deep learning models on unstructured data, including supervised, unsupervised, self supervised learning and transfer learning. The analysis focuses on the required amount and quality of labeled data, support for multimodal inputs, computational cost, and practical applicability in information systems. The study shows that self supervised learning and transfer learning with large pretrained models make it possible to effectively exploit large volumes of unlabeled data and reduce annotation costs while maintaining high solution quality.
unstructured data; machine learning; deep learning; supervised learning; unsupervised learning; self supervised learning; transfer learning; large language models; multimodal models; information systems
1. Ronneberger O. U-Net: Svertochnye seti dlya biomedicinskoy segmentacii izobrazheniy // Medicinskaya komp'yuternaya tomografiya i komp'yuternoe vmeshatel'stvo (MICCAI). 2015. URL: https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/ (data obrascheniya: 13.02.2025).
2. Fyulop A., Horvat A. Skvoznoe obuchenie glubokih neyronnyh setey v chastotnoy oblasti // Mathematics. 2022. T. 10. S. 2132. DOI:https://doi.org/10.3390/math10132132.
3. Lu H., Huan Ya., Chzhan Ch. i dr. Dostizheniya v glubokom obuchenii dlya mikroskopii sverhvysokogo razresheniya (priglashennyy doklad) // Laser & Optoelectronics Progress. 2024. T. 61, № 16. St. 1611002. DOI:https://doi.org/10.3788/LOP6124161002.
4. Mashinnoe obuchenie dlya fizicheskih nauk: vorkshop na NeurIPS 2022 [ML for Physical Sciences Workshop at NeurIPS 2022]. URL: https://ml4physicalsciences.github.io/2022/files/NeurIPS_ML4PS_2022_92.pdf (data obrascheniya: 13.02.2025).
5. Rudskoy A.I., Volkov K.N., Sokolov Yu.A., Kondrat'ev S.Yu. Cifrovye proizvodstvennye sistemy: tehnologii, modelirovanie, optimizaciya. SPb.: SPbPU, 2020. 828 s.
6. Markidis S. Staroe i novoe: Mogut li fizicheski informirovannye metody glubokogo obucheniya zamenit' tradicionnye lineynye reshateli? // Mathematics. 2022. T. 10. S. 603. DOI:https://doi.org/10.3390/math9100603.
7. NVIDIA Solutions. URL: https://www.macnica.co.jp/en/business/semiconductor/manufacturers/nvidia/products/134046/ (data obrascheniya: 13.02.2025).
8. Sizaya A. Modelirovanie teploperedachi i napryazheniy pri lazernoy svarke s ispol'zovaniem paketa FENICS // Angliyskiy yazyk v sfere professional'nyh kommunikaciy: Sb. dokl. VII Vseros. molodezhnoy nauch. konf. Kazan', 2021. S. 249-251.
9. Appercase News. URL: https://www.appercase.ru/news/27220/ (data obrascheniya: 13.02.2025).
10. Assured Systems. URL: https://www.assured-systems.com/nvidia-jetson-agx-xavier-for-new-era-of-ai-in-robotics/ (data obrascheniya: 13.02.2025).
11. Sigma Labs News. URL: https://3dprintingindustry.com/news/sigma-labs-printrite3d-platform-made-compatible-with-polymer-3d-printing-201503/ (data obrascheniya: 13.02.2025).
12. Nikolenko, S. D. Modelirovanie vozniknoveniya vnutrennih napryazheniy v slozhnoy strukture materiala / S. D. Nikolenko, S. P. Kozodaev, S. A. Sazonova // Modelirovanie sistem i processov. – 2024. – T. 17, № 2. – S. 50-61. – DOIhttps://doi.org/10.12737/2219-0767-2024-17-2-50-61.
13. Sazonova, S. A. Modelirovanie processa diagnostiki utechek na osnove dvuhal'ternativnoy gipotezy s uchetom pomeh ot stohastichnosti potrebleniya v gidravlicheskoy sisteme / S. A. Sazonova, I. V. Scherbakova, G. I. Smetankina // Modelirovanie sistem i processov. – 2024. – T. 17, № 1. – S. 111-120. DOI: https://doi.org/10.12737/2219-0767-2024-17-1-111-120
14. Regressionnyy analiz ocenki zagruzki platnogo otdeleniya stomatologicheskoy polikliniki / T. P. Novikova, A. A. Bodin, A. I. Zarevich // Nauchno-tehnicheskiy vestnik Povolzh'ya. – 2023. – № 3. – S. 77–80.
15. Matematicheskaya model' induktivnogo delitelya napryazheniya s elektronnoy kompensaciey / A. I. Zarevich, S. V. Murav'ev // Izvestiya Tomskogo politehnicheskogo universiteta. – 2010. – T. 317, № 4. – S. 129–133.
16. Approximation of the absorption spectrum of indium phosphide in the context of simulation of the process of sensation / Ph. V. Makarenko, V. K. Zolnikov, A. I. Zarevich [et al.] // Russian Microelectronics. – 2024. – Vol. 53, № 4. – P. 329–338. – DOIhttps://doi.org/10.1134/S1063739724040041. DOI: https://doi.org/10.1134/S1063739724600481
17. Formirovanie IKT-kompetentnosti studentov special'nosti «Informatika» / A. V. Poluektov // Informacionno-kommunikacionnye tehnologii v pedagogicheskom obrazovanii. – 2009. – № 3 (3). – S. 8–11. – EDN SAXGJX.
