Early Stopping
/ˈɜːrli ˈstɒpɪŋ/
A regularization technique that stops training when validation performance stops improving.
In italiano: Early StoppingEarly stopping monitors validation loss during training and stops when it hasn't improved for a set number of epochs (patience). This prevents overfitting by stopping before the model memorizes training data.
Examples
- Stop training at epoch 50 when validation loss plateaus
- Patience of 10 epochs
- Restore best weights