Which of the following is used to enhance the amount of training data for a model?

Data augmentation is a technique specifically designed to increase the quantity and diversity of training data without the need to collect new data. It involves creating modified versions of the existing training samples through various transformations, such as rotation, scaling, cropping, or flipping images in the case of visual data. This process helps the model to generalize better by exposing it to a wider range of variations that it might encounter in real-world applications, ultimately improving its robustness and performance.

In contrast, anomaly detection focuses on identifying unusual patterns within data sets, which isn’t directly related to enhancing training data. Dropout is a regularization technique used to prevent overfitting by randomly dropping out neurons during training, while regularization in general is a method used to discourage complex models that may not generalize well. Neither of these methods increases the data available for training; rather, they aim to improve model performance and generalization.