Overfitting
By HubAI editorial team · Source checked: 24 October 2026What does it mean?
Learning training examples so specifically that performance does not generalise well to new data. An impressive training score can conceal weak real-world behaviour.
Picture this
A classifier recognises the invoices used to build it but struggles when a supplier changes its layout. Testing only on the old invoices would hide the problem.
Where you’ll come across it
Model development and evaluation. Keep representative test data separate and revisit it when users, formats or business conditions change.
What it doesn’t mean
A small tidy test set can also mislead. Include edge cases and recent material without quietly turning your test set into another training set.
Compare on a job, not a label
A short exercise · use fictional or non-sensitive material.
- Choose a small set of non-sensitive tasks, including an awkward case and a missing-information case.
- Keep the inputs and scoring checklist the same when you compare two model setups.
- Record useful completions, corrections and elapsed time. Keep the model name and settings with the results.
Is the improvement still useful after you include checking and correction time?
Did that make it clearer?
Likes are a private on-page choice, not a public rating. For a correction or complaint, send the pre-filled form to our editorial inbox. Nothing is sent until you submit it.Check the source
Written and explained by HubAI. The reference below lets you go deeper; it is not an endorsement of a particular product.
Technical reference: developers.google.com ↗