by Michele Laurelli
Layers in a neural network between input and output that learn intermediate representations.
Hidden layers extract increasingly abstract features. Deep networks have many hidden layers. The term 'hidden' refers to not being directly observable as input or output.
Multiple hidden layers in deep networks
Feature extraction layers
Representation learning