Epsilon AI Academy
العربية
Enroll

AI Foundations

How a model learns: under- vs over-fitting

A model that's too simple misses the pattern; one that's too complex memorizes the noise. Add data, add noise, and dial complexity to feel the trade-off that sits underneath all of machine learning.

True pattern Model fit
Good fit
Training error0%
Generalization error0%
40

More examples make over-fitting harder.

10

Real data is messy — noise is what a complex model wrongly memorizes.

4

Polynomial degree. Low = stiff line; high = wild wiggles.

With enough data and the right complexity, the fit tracks the true curve and generalizes.