Сравнение методов

	Онлайн-логистическая регрессия	Регуляризованная логистическая регрессия
Область	Машинное обучение	Машинное обучение
Семейство	Machine learning	Machine learning
Год появления	1960s (perceptron); formalized for logistic loss ~2000s	1996–2005
Автор метода	Rosenblatt, F. / Widrow, B. (perceptron era); modern SGD form: Bottou, L.	Tibshirani, R. (lasso); Hoerl & Kennard (ridge); Zou & Hastie (elastic net)
Тип	Incremental supervised classifier	Penalized classification model
Основополагающий источник	Bottou, L. (2010). Large-Scale Machine Learning with Stochastic Gradient Descent. In Proceedings of COMPSTAT 2010, 177–186. Physica-Verlag.	Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B, 58(1), 267–288.
Другие названия	incremental logistic regression, streaming logistic regression, SGD logistic classifier, online binary classifier	penalized logistic regression, L1 logistic regression, L2 logistic regression, elastic net logistic regression
Связанные	5	5
Сводка	Online Logistic Regression fits a logistic classifier one sample (or mini-batch) at a time via stochastic gradient descent, updating model weights as each observation arrives rather than waiting to see the full dataset. This makes it the standard choice for high-volume, streaming, or memory-constrained binary classification problems where batch training is infeasible.	Regularized logistic regression extends standard logistic regression by adding an L1 (lasso), L2 (ridge), or elastic net penalty to the log-likelihood, shrinking coefficients toward zero and preventing overfitting. It is the default choice for binary or multinomial classification when you want interpretable, sparse, or stable coefficient estimates in high-dimensional or collinear feature spaces.