Small datasets pose significant challenges for training artificial intelligence models. This article analyzes methods aimed at enhancing the generalization capabilities of models, reducing overfitting, and ensuring efficient use of resources. Specifically, transfer learning, meta-learning, Bayesian neural networks, and few-shot learning approaches are comparatively studied. The results indicate that transfer and hybrid methods are effective in increasing accuracy and reducing error on small datasets. The article also provides practical recommendations and directions for future research.
Methods for Improving the Efficiency of Artificial Intelligence on Small Datasets: Alternative Algorithms and Transfer Learning
DOI:
Keywords:
Small dataset
artificial intelligence
transfer learning
Abstract
Downloads
Download data is not yet available.
References
Ўзбекистон Республикаси Президентининг Фармони, 05.10.2020 йилдаги ПФ-6079-сон, «Рақамли Ўзбекистон — 2030» Стратегиясини тасдиқлаш ва уни самарали амалга ошириш чора-тадбирлари тўғрисида.
Oʻzbekiston Respublikasi Prezidentining qarori, 17.02.2021 yildagi PQ-4996-son, Sunʼiy intellekt texnologiyalarini jadal joriy etish uchun shart-sharoitlar yaratish chora-tadbirlari toʻgʻrisida.
Ўзбекистон Республикаси Президентининг 2022 йил 28 январдаги ПФ-60-сон Фармони, 2022-2026 йилларга мўлжалланган Янги Ўзбекистоннинг тараққиёт стратегияси тўғрисида.
Абдуқодиров А.А. Ахборот технологиялари ва таълим. Тошкент, 2020.
Рахматуллаев Х.М. Машинали ўқитиш асослари. Самарқанд, 2021.
Юлдашев Ш.Р. Сунъий интеллект ва таълим. Тошкент, 2022.
B. K. Goh et al., "Machine Learning with Small Datasets," IEEE Access, 2021.
S. J. Pan and Q. Yang, "A Survey on Transfer Learning," IEEE Transactions on Knowledge and Data Engineering, 2010.
K. He et al., "Deep Residual Learning for Image Recognition," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
Y. Gal and Z. Ghahramani, "Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning," in Proceedings of the 33rd International Conference on Machine Learning (ICML), 2016.
C. Finn et al., "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks," in Proceedings of the 34th International Conference on Machine Learning (ICML), 2017.
Y. Wang et al., "Generalizing from a Few Examples: A Survey on Few-Shot Learning," ACM Computing Surveys (CS), 2020.
A. Torralba and A. A. Efros, "Unbiased Look at Dataset Bias," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011.
C. Guo et al., "On Calibration of Modern Neural Networks," in Proceedings of the 34th International Conference on Machine Learning (ICML), 2017.
D. Weiss et al., "A Survey of Negative Transfer," Journal of Machine Learning Research (JMLR), 2022.
A. Kendall and Y. Gal, "What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?," in Advances in Neural Information Processing Systems (NeurIPS), 2017.
J. Snell et al., "Prototypical Networks for Few-shot Learning," in Advances in Neural Information Processing Systems (NeurIPS), 2017.
T. Chen et al., "A Simple Framework for Contrastive Learning of Visual Representations," in Proceedings of the 37th International Conference on Machine Learning (ICML), 2020.
H. Wang et al., "A Survey on Self-Supervised Learning," Artificial Intelligence Review, 2023.