Distinguish between a Weak learner vs Strong Learner

A weak learner refers to a prediction mechanism that produces results that are only slightly more predictive than those resulting from a random chance model. Such a baseline would correspond to a model that simply predicts the overall mean of the target variable in a regression problem or the proportion of success in a binary classification model. In an ensemble based method such as GBM, weak learners typically refer to the simple, shallow decision trees in the first few iterations that eventually get transformed into strong learners through the process of gradient boosting. On the other hand, a strong learner is a highly sophisticated predictive mechanism that is capable of fully understanding the decision boundary of a supervised learning algorithm and thus producing predictions that are significantly more informative than random chance.