Deep learning
Machine learning that uses a big stack of simple maths layers. Stacking many layers lets the program learn complicated patterns, like speech or faces.
A person writes the rules: “If the email mentions ‘lottery’, mark it as spam.”
Show the program ten thousand emails marked spam or not. It works out the patterns itself.
Most of what is called AI today is the second kind. It is powerful because nobody had to write the rules, and hard to explain for exactly the same reason.
The fuller definition
Deep learning uses neural networks with multiple computational layers to learn representations from data.
Go deeper
Lines represent computational connections with learned values. This tiny diagram illustrates a neural network; a transformer or mixture-of-experts model has additional structures not shown here.
A concrete example
Layers can transform text from token representations into patterns useful for predicting a continuation.
The distinction that matters
More layers or parameters do not automatically mean better performance on your task.
Related terms
Neural network
A model made of linked layers of maths. Training adjusts its numbers so it can learn patterns from data.
FoundationsParameter
One of the billions of adjustable numbers inside a model. Training sets them; using the model does not change them.
FoundationsTransformer
The design most modern language models use. Its trick is letting every word in a passage look at every other word to work out what matters.
Where it is taught
What actually happens when you ask AI a question?Source and further reading
Google · Machine Learning GlossaryThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.