This is the book that is only concerned with the theory of deep learning. To explain, from first principles, how realistic deep neural networks operate, tools from theoretical physics are taken and modified. This is to the advantage of both practitioners who are aiming to create better AI models and theorists who are searching for a unified framework to describe intelligence. The field of Artificial Neural Networks is the area that is expanding at the quickest rate within the discipline of Information Technology, and more especially within the fields of Artificial Intelligence and Machine Learning. This essential compendium offers the science behind artificial neural networks as well as case examples of their use. This book is written to assist readers in first understanding the principles and then moving on to polish their programming abilities to become genuine practitioners of deep learning. It covers the fundamental ideas of deep learning as well as the designs of deep learning, such as recurrent neural networks and more recent advancements like generative adversarial networks. In addition, the fundamentals of neural networks and the training of deep neural networks are covered in this book. Students who are seeking a comprehensive reference guide on deep learning should consider this book. Additionally, industry practitioners from a variety of industries who are interested in beginning their journey in the field of data science should also consider this book.