In a feedforward neural network represented as a chain structure, such as $f(x) = f^{(3)}(f^{(2)}(f^{(1)}(x)))$, $f^{(1)}$ is called the first layer, $f^{(2)}$ is the second layer, and so on. The final layer is called the output layer, whose behavior is explicitly specified by the training data to match a label $y \approx f^*(x)$ for each input example $x$. The intermediate layers are called hidden layers because their desired outputs are not explicitly given by the training data. Instead, the learning algorithm must determine how these hidden layers should behave to best implement an approximation of the target function $f^*$.