Hidden LayerRefers to the layers other than the input layer and the output layer in a multi-level feedforward neural network. The hidden layer does not directly receive external signals, nor does it directly send signals to the outside world. It is only needed when the data is separated nonlinearly.

The neurons on the hidden layers can take many forms, such as max pooling layers and convolutional layers, which perform different mathematical functions. If the entire network is viewed as a pipeline of mathematical transformations, the hidden layers are transformed and combined together, mapping the input data to the output space.