Masters Timothy. Deep Belief Nets in C++ and CUDA C. Volume 2. Autoencoding in the Complex Domain

New York: Apress, 2018. — 262 p.This book is a continuation of Volume I of this series. Extensive references are made to material in that volume. For this reason, it is strongly suggested that you be at least somewhat familiar with the material in Volume I.
All techniques presented in this book are given modest mathematical justification, including the equations relevant to algorithms. However, it is not necessary for you to understand the mathematics behind these algorithms. Therefore, no mathematical background beyond basic algebra is necessary to understand the material in this book. However, the two main purposes of this book are to present important deep learning
and data preprocessing algorithms in thorough detail and to guide programmers in the correct and efficient programming of these net algorithms. For implementations that do not use CUDA processing, the language used here is what is sometimes called enhanced C, which is basically C with some of the most useful aspects of C++ without getting into the full C++ paradigm. Strict C (except for CUDA extensions) is used for the CUDA algorithms. Thus, you should ideally be familiar with C and C++, although my hope is that the algorithms are sufficiently clear that they can be easily implemented in any language.
This book is roughly divided into four sections. Chapter 1 presents a technique for embedding class labels into a feature set in such a way that generative exemplars of the classes can be found. Chapters 2 and 3 present signal and image preprocessing techniques that provide effective inputs for deep belief nets. Special attention is given to preprocessing that produces complex-domain features. Chapter 4 discusses basic autoencoders, with emphasis on autoencoding entirely in the complex domain. This is particularly useful in many fields of signal and image processing. Chapter 5 is a reference for the DEEP program, available as a free download from my web site.