Skip to content

Model Training & Adaptation

Self-training

AI using its own guesses about new examples as material for further learning.

Example

AI adds selected guesses about new examples to its teaching material.

Why people use it

It can expand the examples available for learning by reusing selected predictions.

What you'll hear

“Use the strongest guesses for another learning round.”

What this means for you

Use independent checks between rounds so mistakes are not repeatedly reinforced.

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 self-training reinforce existing mistakes?
Yes. Incorrect predictions can become misleading training targets.
Can repeated rounds reduce variety?
Yes. Favoring familiar, easy-to-label cases can leave uncommon cases behind.
Does self-training mean no people are involved?
No. People still choose the information, method and checks around the process.

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