What is backpropagation?

Backpropagation is an algorithm for training neural networks. It is called so because of the way it works: iterating from the last layer to the first one (backward). Its primary purpose is to update the weights of the neural network by calculating the gradient of the loss function.

Why is it important?

Backpropagation is a commonly used method for training feedforward neural networks. It is an efficient algorithm, which allows us to work with large networks with many neurons.