Neural networks
An artificial neuron is the basic building block of a neural network: a small unit that takes in numbers, weighs them, adds them up, and sends out a signal.
Each neuron does one simple job. It receives several input numbers, multiplies each by its own weight, which says how much that input matters, and adds the results. Then it passes the total through an activation function that decides how strongly to 'fire.' That output becomes an input for neurons in the next layer.
A single neuron can only learn very simple things, like whether a weighted mix of inputs crosses a line. The power comes from stacking. Millions of neurons across many layers combine their tiny decisions into rich judgments, such as recognizing a word or a face.
Despite the name, an artificial neuron is nowhere near a biological one. A real brain cell is a living thing with complex chemistry. The artificial version is a few lines of arithmetic. Researchers borrowed the word in the 1940s, and it stuck.
In an image model, one neuron might grow sensitive to a curved edge in the top left of a picture, firing strongly whenever that shape appears.