Single Neuron

A neuron is the fundamental unit of a neural network, responsible for processing information. It receives input data (any data encoded as numerical values), processes it, and passes the result forward. Each input is assigned a weight, which determines its importance in the neuron's calculation.

The way each neuron works can be broken down into 4 steps:

1. Receiving input: a neuron takes multiple inputs, represented as $x_1$, $x_2$, $x_3$, etc;
2. Applying weights: each input is multiplied by a corresponding weight $w_1$, $w_2$, $w_3$, etc., which defines its significance. Initially, these weights are set randomly, but they are later adjusted during training using a method called backpropagation, which we will cover later;
3. Summation: the weighted inputs are summed together ($w_1 \times x_1 + w_2 \times x_2 + ...$);
4. Activation function: this sum is passed through an activation function, which transforms it into the neuron's output. The choice of activation function depends on the problem the neural network is trying to solve.

Neuron as Part of a Neural Network

The neuron's output serves as input for the next layer of neurons. This process continues through multiple layers until the network produces a final result.

During the learning process, the weights of the neuron are adjusted in such a way as to minimize the error between the values predicted by the neural network and the real values. .

If the network makes a mistake, it updates the weights to improve future predictions.

By refining these connections over time, the neural network learns patterns in data and improves its accuracy in solving tasks.