Glossary
What is PINN? Definition and meaning
A neural network trained with the governing physical equations as part of its loss, so predictions respect the physics.
Definition
PINN: A neural network trained with the governing physical equations as part of its loss, so predictions respect the physics.
In short: Physics-informed neural network.
When it is used
Surrogate models and digital twins of physical processes with limited data.
Learn more
The full guide to PINN explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Pinecone · Next term: Pipeline parallelism