What is early stopping in supervised learning?
Updated May 17, 2026
Short answer
Early stopping halts training when validation performance stops improving.
Deep explanation
Early stopping prevents overfitting by monitoring validation loss during training. When performance plateaus or worsens, training is stopped. It is widely used in neural networks and boosting models.
Real-world example
Training deep learning models for image recognition without overfitting noise.
Common mistakes
- Monitoring training loss instead of validation loss.
Follow-up questions
- What is patience in early stopping?
- Why is early stopping effective?