What is overfitting and underfitting?

Overfitting and underfitting are common problems in Machine Learning that affect how well a model performs.

Overfitting happens when a model learns the training data too well, including noise and small details that don’t generalize to new data. As a result, the model performs very well on training data but poorly on unseen data. Think of it like memorizing answers for an exam instead of understanding the concepts — you do well on familiar questions but struggle with new ones.

Underfitting happens when a model is too simple to capture the underlying patterns in the data. It fails to learn properly, so it performs poorly on both training data and new data. This is like not studying enough — you don’t perform well even on basic questions.

In short:

• Overfitting = too complex, memorizes data, poor **generalization**
• Underfitting = too simple, fails to learn patterns

A good model finds the right balance, where it learns meaningful patterns without **memorizing noise.**