手法を比較

	ロバストオンライン学習	ロバスト勾配ブースティング
分野	機械学習	機械学習
系統	Machine learning	Machine learning
提唱年	2000s–2010s	2001
提唱者	Hazan, E.; Shalev-Shwartz, S.; and others	Friedman, J. H. (with Huber loss from Huber, P. J.)
種類	Algorithmic framework	Ensemble (boosted trees with robust loss)
原典	Hazan, E. (2016). Introduction to Online Convex Optimization. Foundations and Trends in Optimization, 2(3–4), 157–325.	Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 29(5), 1189–1232.
別名	ROL, robust incremental learning, adversarially robust online learning, robust sequential learning	gradient boosting with Huber loss, robust GBM, outlier-robust boosting, robust gradient-boosted trees
関連	5	6
概要	Robust Online Learning extends the online learning framework — where a model updates sequentially after each observation — by incorporating robustness mechanisms that guard against corrupted labels, adversarial examples, heavy-tailed noise, and concept drift. The result is a sequential learner that maintains bounded regret even when the data stream contains outliers or deliberate perturbations.	Robust Gradient Boosting is gradient boosting trained with outlier-resistant loss functions — most commonly the Huber loss or quantile (pinball) loss — instead of squared-error loss. Proposed in Friedman's seminal 2001 paper, this variant produces predictions far less distorted by extreme values or contaminated labels, while retaining the full predictive power of gradient-boosted trees.