Introduction to Deep Learning: Mathematical Foundations and Hands-on Implementations

by Jai Sharma

In this text, we will start by exploring the applications of Deep Learning and why it has become so widespread in its use. We will then examine some elementary Machine Learning algorithms such as Linear Regression and Logistic Regression, both of which will give us a nice foundation to learn about Neural Networks. Next, we will explore Neural Networks at a high level without going too much into the mathematical side.

After this high level approach, we delve into the basic mathematical understanding needed for Deep Learning, including Linear Algebra and Calculus. We develop the tools we will need to derive the parameter updates for Linear Regression, Logistic Regression, and Neural Networks. These derivations are a essential to explaining how Machine Learning and Deep Learning algorithms “learn” from patterns.

The second part of the text focuses on implementing the skills in the first part in code. We will be using Python here because of its famous libraries such as NumPy and Matplotlib. We will conclude the text with implementations of Linear Regression, Logistic Regression, and Neural Networks from scratch in Python.

I hope this book will not only provide instruction on the concepts behind Deep Learning, but also the interest to explore its riches. Thank you for reading this text.

167 pages, Kindle Edition

Published November 13, 2021