[TBD]Role of data augmentation in machine learning

Introduction

Laymen explanation

Technical explanation

Jargons

White noise

White Noise, by definition, works by defining parameters in which data is ensured to be random, unrelated, and have zero mean.

Desirable properties

• Modifications made for generating new data should be learnable. Simply adding white noise will not help. Note that white noise is not learnable.

Techniques

Gaussian Noise

Reference

https://www.amazon.in/Hands-Machine-Learning-Scikit-Learn-TensorFlow/dp/1491962291

https://deepai.org/machine-learning-glossary-and-terms/white-noise

https://nanonets.com/blog/data-augmentation-how-to-use-deep-learning-when-you-have-limited-data-part-2/