A neural network is the core structure behind most modern AI. It is made of layers of simple connected units ("neurons") that each pass signals forward. During training, the strengths of those connections are adjusted over and over until the network reliably turns inputs into the right outputs — turning the pixels of a photo into the label "dog," for example.
The name comes from a loose analogy with the brain, but a neural network is really just mathematics and adjustable numbers. When a network has many layers, the approach is called deep learning, which is what made today's powerful language and image models possible.
Related terms
Machine Learning (ML) · Training Data · Large Language Model (LLM)