방법 비교

	그래디언트 부스팅	온라인 배깅	온라인 학습
분야	머신러닝	머신러닝	머신러닝
계열	Machine learning	Machine learning	Machine learning
기원 연도	2001	2001	1958–2000s
창시자	Friedman, J. H.	Oza, N. C. & Russell, S.	Rosenblatt, F.; Littlestone, N.; Shalev-Shwartz, S. (key contributors)
유형	Ensemble (sequential boosting of decision trees)	Online ensemble (streaming bagging)	Learning paradigm (sequential model update)
원전	Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. Annals of Statistics, 29(5), 1189–1232.	Oza, N. C., & Russell, S. (2001). Online bagging and boosting. In Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics (AISTATS 2001), pp. 105–112.	Shalev-Shwartz, S. (2011). Online Learning and Online Convex Optimization. Foundations and Trends in Machine Learning, 4(2), 107–194.
별칭	Gradient Boosting (GBM), GBM, gradient boosted trees, gradient boosting machine	incremental bagging, streaming bagging, online bootstrap aggregating, OzaBag	incremental learning, sequential learning, streaming learning, online machine learning
관련	5	4	6
요약	Gradient Boosting is an ensemble learning method, formalised by Jerome H. Friedman in 2001, that combines a sequence of weak learners — typically shallow decision trees — so that each new tree is fitted to minimise the residual errors of the trees before it. It is the core algorithm behind popular implementations such as XGBoost, LightGBM and CatBoost.	Online Bagging is a streaming ensemble method introduced by Oza and Russell in 2001 that adapts the classical bootstrap aggregating (Bagging) framework to the online learning setting. Instead of resampling a fixed dataset, each incoming instance is fed to every base learner a Poisson(1)-distributed number of times, faithfully approximating bootstrap sampling as the stream evolves. The result is a robust, incrementally updated ensemble that can handle concept drift and continuous data arrival without storing the entire dataset.	Online learning is a machine learning paradigm in which a model is updated incrementally as each new data point arrives, rather than being trained once on a fixed dataset. It is essential when data streams continuously, storage is limited, or the underlying distribution shifts over time. Theoretical performance is measured by cumulative regret relative to the best fixed predictor in hindsight.