Overfitting

AI/ML Fundamentals

Model learns training data too well, performs poorly on new data

What is Overfitting?

When a model memorizes training data including noise rather than learning general patterns. Like a student who memorizes answers but can't solve new problems.

Real-World Examples

•Model achieves 99% accuracy on training but 60% on test data
•Decision tree with too many branches

Common Mistakes to Avoid

Using overly complex models or too little training data

Related Terms

Learn more about concepts related to Overfitting

Underfitting

Model is too simple to capture patterns in data

Regularization

Techniques to prevent overfitting

Validation

Evaluating model performance during training