Skip to content

AI Foundations

ReLU

Rectified linear unit

A calculation that keeps positive numbers unchanged and turns negative numbers into zero.

Example

An AI stage changes a negative number to zero while leaving a positive number as it is.

Why people use it

It gives networks a simple rule that helps them learn more than straight-line relationships.

What you'll hear

“Use this simple rule inside the network.”

What this means for you

Treat this as one small part of the system, not a measure of its overall ability.

Can you control it?

Developer-only

The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.

Common questions

Can this calculation lose information?
Yes. Different negative numbers all become zero, so their original differences are no longer visible afterward.
What happens to a negative value?
ReLU turns it into zero rather than passing the negative value onward.
Can a unit become inactive?
Yes. If it keeps receiving negative values that become zero, it can stop contributing useful learning.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice