Data Augmentation

A technique used to increase the diversity and size of a training dataset by creating modified copies of existing data or generating new synthetic data. For images, this might include rotating, cropping, or changing the brightness. This helps improve model performance and reduce overfitting.