Бустинг

Самченко, С. О.

Публікація:
Бустинг

dc.contributor.author	Самченко, С. О.
dc.date.accessioned	2024-08-23T07:14:26Z
dc.date.available	2024-08-23T07:14:26Z
dc.date.issued	2024
dc.description.abstract	Boosting is a powerful machine learning technique that combines multiple weak models to create a strong ensemble model. It is based on the idea of iterative learning, during which each subsequent model focuses on the mistakes made by previous models. One of the most common boosting options is AdaBoost (Adaptive Boosting), developed by Freund and Schapire in 1996. This method has high accuracy on complex problems and is robust to overfitting, which allows its wide application in various fields such as classification and regression. AdaBoost uses error weighting to update the weights of the samples in the training set, which improves its efficiency. In addition to AdaBoost, there are other popular boosting algorithms, such as Gradient Boosting, XGBoost, LightGBM and CatBoost, which are also successfully used in practice. It is important to properly configure boosting parameters to achieve optimal model performance.
dc.identifier.citation	Самченко С. О. Бустинг / С. О. Самченко ; наук. керівник О. Є. Путятiна // Радіоелектроніка та молодь у XXI столітті : матеріали 28-го Міжнар. молодіж. форуму, 16–18 квіт. 2024 р. – Харків : ХНУРЕ, 2024. – Т. 2. – С. 45–46.
dc.identifier.uri	https://openarchive.nure.ua/handle/document/27948
dc.language.iso	uk
dc.publisher	ХНУРЕ
dc.subject	boosting
dc.subject	бустинг
dc.title	Бустинг
dc.type	Thesis
dspace.entity.type	Publication