A popular boosting algorithm and software library (stands for "extreme gradient boosting"). Boosting combines weakly predictive models into a strongly predictive model.

XGBoost stands for Extreme Gradient Boosting. It is an algorithm (Chen & Guestrin, 2016) for gradient boosting decision trees. XGBoost adds several features beyond existing algorithms for boosting trees: it is optimized for sparse data and it uses the data's block structure. This makes it faster than previous algorithms. It has been implemented in an R package.

  • Tianqi Chen and Carlos Guestrin. 2016. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). ACM, New York, NY, USA, 785-794. DOI: https://doi.org/10.1145/2939672.2939785
history | excerpt history