Skip to content

Model Training & Adaptation

Semi-supervised learning

Teaching AI with a mixture of examples that have answers attached and examples that do not.

Example

AI learns from some labeled pictures alongside many pictures without labels.

Why people use it

It makes use of a large collection when only some examples have checked answers.

What you'll hear

“Only a small part of this collection has been labeled.”

What this means for you

Compare results with learning from the labeled examples alone.

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

Does unlabeled data always improve the result?
No. Incorrect assumptions or poor data can reduce performance.
Can the unlabeled portion describe a different kind of case?
Yes, and that mismatch can make the additional material less helpful.
Does the method still need a final independent check?
Yes. Learning from more material does not establish how well it handles genuinely new cases.

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