Computational Methods for Deep Learning: Theoretic, Practice and Applications

Description

In this book, we work for the contents for knowledge transfer from the viewpoint of machine intelligence. We adopt the methodology from graphical theory, mathematical models, algorithmic implementation as well as datasets preparation, programming, results analysis and evaluations. We start from understanding artificial neural networks with neurons and the activation functions, then explain the mechanism of deep learning using advanced mathematics. We especially emphasize on how to use TensorFlow and the latest MATLAB deep learning toolboxes for implementing deep learning algorithms. Before reading this book, we strongly encourage our readers to understand the knowledge of mathematics, especially those subjects like mathematical analysis, linear algebra, numerical analysis, optimizations, differential geometry, manifold, information theory as well as basic algebra, functional analysis, graphical models, etc. The computational knowledge will assist us not only in understanding this book and but also in relevant deep learning journal articles and conference papers. This book was written for research students and engineers as well as computer scientists who are interested in deep learning for theoretic research and analysis. More generally, this book is also helpful for those researchers who are interested in machine intelligence, pattern analysis, natural language processing, and machine vision.