Which of the following is NOT a feature of backpropagation?

Group of answer choices

Backpropagation is a common method of artificial neural network training.

Backpropagation reduces the dimensionality of the network.

Backpropagation calculates the error from the end to the beginning of the network.

Backpropagation calculates the gradient $($the change between the values of each node$)$ layer by layer.