Overfitting
Domain: Evaluation
🎯 Motivation
The model memorizes training data but cannot generalize to new data. Like a student who memorizes answers without understanding.
📋 Concrete Example
Overfit model: 99% on training data, 60% on test data. It memorized specific examples but cannot solve new problems. Prevention: regularization, more data, early stopping.
🔗 Analogy
A student who memorizes past exam answers - they score 100% on old exams but fail the real exam with new questions. Overfitting is memorizing without understanding.