GENERATING SYNTHETIC DATA FOR TRAINING ARTIFICIAL INTELLIGENCE
Keywords:
synthetic data, data generation, artificial intelligence, generative models, GANAbstract
This scientific article provides a comprehensive analysis of the generation of
synthetic (artificially created) data for training artificial intelligence (AI) systems and the prospects
of this direction. Modern AI models require enormous amounts of data for training, but collecting
real data is often expensive, time-consuming, or impossible due to privacy concerns. The research
scientifically substantiates the methods of creating artificial data that resembles real data (generative
models, simulations), its advantages, and its limitations. The article examines the use of synthetic
data in solving privacy problems, eliminating data shortages, and balancing datasets. The scientific
novelty of the article lies in demonstrating that synthetic data is becoming an important tool for the
development of AI. As a result of the analyses, recommendations are developed regarding the
application of these technologies and their reliability
References
1. Goodfellow, I., et al. (2014). "Generative Adversarial Nets." Advances in Neural Information
Processing Systems.
2. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.
3. Nikolenko, S. I. (2021). Synthetic Data for Deep Learning. Springer.
4. Jordon, J., et al. (2019). "PATE-GAN: Generating Synthetic Data with Differential Privacy
Guarantees." ICLR.
5. Bowles, C., et al. (2018). "GAN Augmentation: Augmenting Training Data using Generative
Adversarial Networks." arXiv preprint.
6. Choi, E., et al. (2017). "Generating Multi-label Discrete Patient Records using Generative
Adversarial Networks." Machine Learning for Healthcare.
7. Shorten, C., & Khoshgoftaar, T. M. (2019). "A survey on Image Data Augmentation for Deep
Learning." Journal of Big Data.
8. Vaswani, A., et al. (2017). "Attention is All You Need." Advances in Neural Information
Processing Systems.
9. Dwork, C., & Roth, A. (2014). "The Algorithmic Foundations of Differential Privacy."
Foundations and Trends in Theoretical Computer Science.
10. Tremblay, J., et al. (2018). "Training Deep Networks with Synthetic Data: Bridging the Reality
Gap by Domain Randomization." CVPR Workshops.
11. Xu, L., et al. (2019). "Modeling Tabular Data using Conditional GAN." Advances in Neural
Information Processing Systems.
12. Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach (4th ed.). Pearson.
13. Kingma, D. P., & Welling, M. (2014). "Auto-Encoding Variational Bayes." ICLR.
14. Decree of the President of the Republic of Uzbekistan "On the approval of the strategy 'Digital
Uzbekistan – 2030'" (October 5, 2020).
15. Assefa, S. A., et al. (2020). "Generating Synthetic Data in Finance: Opportunities, Challenges
and Pitfalls." ACM International Conference on AI in Finance
16. UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence.
17. Law of the Republic of Uzbekistan "On Personal Data" (2019).
18. Bostrom, N. (2014). Superintelligence: Paths, Dangers, Strategies. Oxford University Press.
