What is overfitting and how do you prevent it?

Overfitting occurs when models learn training data too well, including noise, resulting in poor generalization to new data.\n\n• Detection: High training accuracy but low validation accuracy\n• Prevention: Cross-validation, regularization (L1/L2), early stopping, dropout\n• Data approaches: More training data, data augmentation, feature selection\n• Model complexity: Simpler models, ensemble methods, hyperparameter tuning\n\nExample: Deep learning model for image classification shows 99% training accuracy but 70% validation accuracy. Apply dropout layers, reduce model complexity, use data augmentation, implement early stopping, and regularization to improve generalization.