In the context of gradient boosting, weak learners are individual models that perform slightly better than random guessing but are not highly accurate on their own. Decision trees are commonly used as weak learners in gradient boosting, but other models can also be used.

Each connection between neurons is assigned a weight, which represents the strength or importance of that connection. These weights determine how much influence the input of one neuron has on the output of another.