방법 비교

	온라인 LightGBM	그래디언트 부스팅	온라인 학습
분야	머신러닝	머신러닝	머신러닝
계열	Machine learning	Machine learning	Machine learning
기원 연도	2017 (LightGBM); 2000s (online boosting)	2001	1958–2000s
창시자	Ke et al. (LightGBM); Bifet, Gavalda (online boosting theory)	Friedman, J. H.	Rosenblatt, F.; Littlestone, N.; Shalev-Shwartz, S. (key contributors)
유형	Online ensemble (incremental gradient boosting)	Ensemble (sequential boosting of decision trees)	Learning paradigm (sequential model update)
원전	Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A Highly Efficient Gradient Boosting Decision Tree. Advances in Neural Information Processing Systems, 30.	Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. Annals of Statistics, 29(5), 1189–1232.	Shalev-Shwartz, S. (2011). Online Learning and Online Convex Optimization. Foundations and Trends in Machine Learning, 4(2), 107–194.
별칭	Incremental LightGBM, LightGBM incremental training, streaming LightGBM, continual LightGBM	Gradient Boosting (GBM), GBM, gradient boosted trees, gradient boosting machine	incremental learning, sequential learning, streaming learning, online machine learning
관련	5	5	6
요약	Online LightGBM applies the Light Gradient-Boosting Machine framework incrementally: instead of requiring all training data at once, the model is updated in mini-batches or data chunks as they arrive. This allows LightGBM's efficient histogram-based boosting to be deployed in streaming, continual-learning, and data-expansion scenarios without retraining from scratch.	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 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.