Hands-On Automated Machine Learning

Neuron

Neurons are the basic unit of a neural network. Each neuron has a number of inputs, a processor, and an output. The processing of a neuron starts by accumulating all of its incoming inputs, and, based on the activation function, it fires a signal that is propagated to other neurons or as an output signal. This helps neurons to learn. The most basic form of a neuron is known as the perceptron. A perceptron has a binary output of either 0 or 1.

A typical neuron looks like the following:

As we can see, neurons have several inputs. For each input connection, there is a weight associated with that specific connection. When the neuron is ...