Neural Network
A model built from layers of simple connected units ('neurons') that transform an input into an output, learning the connection strengths from data. It's the workhorse behind modern AI.
Think of It Like This
A relay team of simple runners — no single one is clever, but the handoffs add up to something powerful.
Each neuron takes weighted inputs, adds them up, and fires through an activation function; stack enough of these into layers and the network can approximate almost any pattern. Training adjusts the connection weights with gradient descent so the output error steadily shrinks. Everything from image classifiers to large language models is, at heart, a neural network — the multi-layer perceptron is the simplest version to start with.