AI Terms Dictionary

A comprehensive multilingual AI terminology dictionary

Definition

Early stopping is a form of regularization used primarily in iterative training processes like gradient descent. During training, the model’s performance on the training data typically improves continuously, but its ability to generalize to unseen data may start to decline after a certain point, indicating overfitting. Early stopping monitors a validation metric; if this metric fails to improve for a predefined number of epochs (patience), training is terminated. The model weights from the best-performing epoch are then restored. This technique effectively selects the optimal complexity of the model without requiring explicit penalty terms in the loss function.

Summary

Early stopping is a regularization technique that halts the training process when the model’s performance on a validation set begins to degrade, preventing overfitting.

Key Concepts

Use Cases