Difference between a "node" and an "input" in a neural network

Question

I am reading an article which is opened with the sentence:

We concider a 2-layer, 3-node, n-input neural network whose nodes compute linear threshold functions of their inputs.

I don't understand how is it possible for a neural network to have 3 nodes but n inputs. For my understanding, a "neuron" is a "node". So there're n input nodes (first layer) and x output nodes (second layer) and the sentence arguing n+x=3.

What do I miss? Thanks!

Answer 1

Look at this 2-layer NN in the picture. This has:

• 3 inputs
• 2-layer with 4 nodes
• and 1 output

When we say "an n-layer NN" we are counting its hidden layers (because obviously it has input and output so don't need to be count).