Which of the following statements are true for bagging? $($may have multiple correct answers, you

will be penalized from wrong answers$)$

a$.$ The main purpose of bagging is to decrease the bias of learning algorithms.

b$.$ In bagging, we choose random subsamples

c$.$ Bagged trees might be correlated.

d$.$ Bagging is ineffective with logistic regression, because all of the learners learn exactly the

same decision boundary.