PERFORMANCE AND OPTIMIZATION METHODS OF ARTIFICIAL INTELLIGENCE MODELS WHEN WORKING WITH DATA

This thesis deeply explores the principles of artificial intelligence (SI) models in data processing and their optimization methods. The ability of modern SI models to work with various types of data, including large-volume and complex structured data processing processes, is analyzed in detail. In the thesis, a comparative analysis of model training algorithms, in particular, deep learning and machine learning methods, as well as modern approaches and techniques used to optimize their performance, is widely discussed. The results of the research show the possibilities of increasing the effectiveness of SI models and expanding their practical application in various fields.