Data augmentation techniques have been used very successfully to increase the amount of training data, most notably in image processing. Augmented training set reduces overfitting when training an ML model by acting as a regularizer. In this talk, we will explore multiple solutions for augmenting tabular data.