See Boosting for Regression. In classification, we repeatedly apply a weak binary classifier to the training set while modifying sample weights.

We take a look at the AdaBoost.M1 Algorithm as an example of Boosting for Classification.