Data Augmentation

Definition

Data augmentation enhances machine learning model robustness by artificially expanding the training dataset through transformations like rotation or scaling, enabling better generalization capabilities despite limited original data availability.

When Data Augmentation is used

Image rotation or flipping

Which positions need this?

Data Scientists

Problem

Overfitting due to small training datasets.

Example of how Data Augmentation is used in AI

A computer vision project employs data augmentation techniques like random cropping and flipping to enhance the dataset size from hundreds to thousands of unique images for model training.