Skip to content

Model Training & Adaptation

Bias-variance tradeoff

The tension between AI being too simple to learn useful patterns and too flexible to ignore accidental ones.

Example

A small AI system underfits while a complex one performs poorly on new cases.

Why people use it

It helps explain why both overly simple and overly complicated predictors can fail.

What you'll hear

“Are we missing the pattern or learning the noise?”

What this means for you

Choose complexity using performance on representative unseen data.

Can you control it?

No

No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.

Common questions

Is bias in this phrase the same as social unfairness?
No. Here it refers to systematic modeling error, although fairness is a separate concern.
Can a more complex system do worse on new examples?
Yes. It can learn accidental details that do not carry over to new situations.
Does one perfect training score settle the tradeoff?
No. Performance on unfamiliar examples is needed to judge whether the learned pattern is useful.

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