What does overfitting mean?

When a model fits its training data too closely, noise included, so that it does well on the data it was fitted to and badly on new data. It is found by testing on held-out data the fit never saw.

Also written as: overfit, overfits, overfitted

In a lesson

Sample fourteen commands, six for training and eight for testing, fit a degree 1 and a degree 5 model to the training six, print all four errors, and print train 5: and best degree:. Three of the test commands, 25, 85 and 95, lie outside the training range, which is where an overfitted model is at its worst.

Learn it

• Taught in U12.3 Generalisation Learning, and the capstone, University

All glossary terms