Skip to content

Model Training & Adaptation

Self-supervised learning

AI learning from practice tasks made from the material itself, rather than answers separately supplied by people.

Example

An AI system learns to predict hidden words in a passage.

Why people use it

It allows learning from large collections without people labeling every example by hand.

What you'll hear

“The system makes its own practice questions from the text.”

What this means for you

Distinguish automatic target creation from the absence of oversight.

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 predicting missing words the only practice task?
No. Other approaches learn from relationships between pictures, sounds or different parts of the available material.
Does that mean it learns without any human choices?
No. People still choose the material, learning task and way the system is built.
Can it learn harmful patterns too?
Yes. Learning automatically from material does not remove problems already present in that material.

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