Skip to content

Model Training & Adaptation

Machine unlearning

Methods that try to remove or reduce what an AI learned from specified information.

Example

A team checks whether AI can lose the influence of a selected set of examples.

Why people use it

It aims to reduce a learned influence without simply rebuilding everything from the beginning.

What you'll hear

“Can the system forget these teaching examples?”

What this means for you

Ask how removal was tested and what evidence of remaining influence was checked.

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

Is deleting the original file the same as unlearning it?
No. Information may remain reflected in learned learned settings.
Is success easy to prove?
Not always. A system may stop repeating a fact while other traces of its influence remain.
Can removing one influence affect other abilities?
It can. Learned information can be connected to other patterns the system uses.

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