SELF-LEARNING OF ARTIFICIAL INTELLIGENCE. BASIC PRINCIPLES OF ARTIFICIAL INTELLIGENCE OPERATION USING A SIMPLE EXAMPLE
Sadikova M. A. Fergana Branch of TUIT named after Muhammad al-Khwarizmi, Senior Lecturer at the Department of Software Engineering and Digital Economy. Avazova N. K. 3rd-year student at the Faculty of Software Engineering and Digital Economy.
Keywords:
Искусственный интеллект, машинное обучение, самообучение, интеллектуальные системы, автономная навигация, переобучение, стабильные алгоритмы, базовые принципы, нейронные сети, вероятные метрики, набор данных MNIS, гиперпараметры, глубокое обучение, валидация моделиAbstract
Annotation: The article provides a comprehensive overview of modern concepts, methods, and the significance of self-learning for the development of AI. Starting with the definition and technical aspects of self-learning, the article delves into the core advantages and challenges of this approach, emphasizing effective strategies to manage constraints. Additionally, it conducts research that illuminates the fundamental principles of neural network operation. The training results of the model are depicted in the change of accuracy on both training and test data in each epoch, allowing for an assessment of the model's performance and identification of necessary enhancements to improve its efficiency.
References
"Глубокое обучение" - Иэн Гудфеллоу, Йошуа Бенджио, Аарон Курвилль. MIT Press, 2016.
"Обучение с подкреплением: Введение" - Ричард С. Саттон, Эндрю Г. Барто. The MIT Press, 2018.
"Life 3.0: Being Human in the Age of Artificial Intelligence" - Макс Тегмарк. Penguin Books, 2018.
"Искусственный интеллект: современный подход" - Стюарт Рассел, Питер Норвиг. Pearson, 2021.
"Сверхинтеллект: пути, опасности, стратегии" - Ник Бостром. Oxford University Press, 2016.
https://colab.research.google.com/drive/1S8CYaNna7EGmRBSLGWdr95hLH5oFjCaF?usp=sharing Ссылка на оригинальную работу
Additional Files
Published
How to Cite
License
Copyright (c) 2023 Navruzakhon Avazova, М.А. Садикова

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.