Deep learning
Machine learning with many stacked layers, so the model learns its own features instead of being handed them.
Think of it like
Teaching someone to cook by letting them ruin a thousand dishes, rather than writing out every rule for every ingredient.
Example
Image recognition jumped from hand-crafted edge detectors to deep learning around 2012, and never looked back.
How it actually works
The “deep” is literal: many layers between input and output. Each layer builds on the last, so the network discovers useful representations on its own from raw data. It’s expensive and data-hungry, but it scales in a way older methods didn’t.
For product teams
It’s the engine under nearly every modern AI product — the reason the last decade happened.
For engineers
Neural networks with many hidden layers trained by backpropagation and gradient descent.
Related
- Machine learning — A branch of machine learning.
- Layer — Built from stacked layers.
- Neural network — Runs on a neural network.
Read anything AI without the jargon
Look up any term in plain English, or save terms as you read with the free Chrome extension.
Open DecoderAdd to Chrome