Neural Networks & Deep Learning
Concept
Core Technical
The substrate: layered functions tuned by gradient descent.
You're looking at a concept explainer: what this term means in plain English, and the records it ties together.
A neural network is a stack of simple mathematical layers whose millions to trillions of weights are adjusted by backpropagation until outputs match training targets. 'Deep learning' is the discovery — mainstream since AlexNet in 2012 — that with enough layers, data and GPU compute, this recipe learns almost anything. Everything else on this site is scaffolding around that fact.
How this connects to the rest of the stack
This record's links at a glance — click a group to jump to it.
Related
Sources
Record reviewed as of 2026-01-15 — see Methodology.