Skip to content

Model Training & Adaptation

Hyperparameter tuning

Comparing different setup choices to find ones that help an AI perform well.

Example

A team compares learning rates on a independent checking set.

Why people use it

It finds useful learning settings through comparison rather than guesswork.

What you'll hear

“Which settings worked best on the practice checks?”

What this means for you

Avoid choosing settings from repeated inspection of the final test set.

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 the best settings change later?
Yes. New information, different tasks or changed computing limits can make other settings more suitable.
Can settings that win on one collection lose on another?
Yes. The best choice can depend on the task and the examples used.
Why keep a final test separate?
Repeatedly choosing settings from the same test can make the final score look better than real performance.

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