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?

More Supervised Learning interview questions

View all →