Artificial neural networks are typically arranged in a topology of several layers.
The input neurons are in the first layer and the output neurons are in the
last. Additional layers of neurons (called hidden layers) may be included
between the input and output layers. Each neuron of one layer is interconnected
with every neuron in the subsequent layer. As an example, the simple network
presented in Figure 11.18a is programmed to produce an output of 1 if its two
inputs differ and an output of 0 otherwise. If, however, we change the weights to
those shown in Figure 11.18b, we obtain a network that responds with a 1 if both
of its inputs are 1s and with a 0 otherwise.
We should note that the network configuration in Figure 11.18 is far more
simplistic than an actual biological network. A human brain contains approximately
1011 neurons with about 104 synapses per neuron. Indeed, the dendrites of
a biological neuron are so numerous that they appear more like a fibrous mesh
than the individual tentacles represented in Figure 11.15.