XGBoost (eXtreme Gradient Boosting) is an open-source machine learning library that implements gradient boosting, an algorithm that reduces prediction error by sequentially adding decision trees. It was developed by Tianqi Chen at the University of Washington and released in 2014, adding regularization, parallel processing, and missing-value handling to improve speed and accuracy. It is available for Python, R, Java, C++, and other languages.
It became widely known through its frequent use in winning entries in data science competitions such as Kaggle and is a standard tool for classification, regression, and ranking on tabular data. Together with LightGBM and CatBoost, it is considered one of the leading gradient boosting libraries.