دانلود کتاب روش های محاسباتی برای یادگیری عمیق

This book has been drafted based on my lectures and seminars from recent years for postgraduate students at Auckland University of Technology (AUT), New Zealand. We have integrated materials on deep learning, machine learning, and computational mathematics, refining the contents to publish this book. Our aim is to provide a resource that benefits postgraduate students, particularly those working on their theses, by sharing our research outputs and teaching work to enhance their projects. In this book, we organize our material and present the story of deep learning in a progression from easy to difficult concepts in mathematics. We have structured the content with a focus on knowledge transfer from the perspective of machine intelligence. We begin by explaining artificial neural networks, including neuron design and activation functions. We then delve into the mechanics of deep learning using advanced mathematical concepts. At the end of each chapter, we emphasize the practical implementation of deep learning algorithms using Python-based platforms and the latest MATLAB toolboxes. Additionally, we provide a list of questions for reflection and discussion. Before reading this book, we strongly encourage our readers to have a solid foundation in postgraduate mathematics, including subjects such as basic algebra, functional analysis, graphical models, and other fundamental topics like mathematical analysis, linear algebra, probability theory, mathematical statistics, optimization theory, computational methods, differential geometry, manifold, and information theory. Developing computational knowledge will not only help readers understand this book but also enable them to engage with relevant journal articles and conference papers in the field of deep learning. This book is written for research students, engineers, computer scientists, and anyone interested in computational methods of deep learning for both theoretical analysis and practical applications. Additionally, it is relevant for researchers in fields such as machine intelligence, pattern analysis, computer vision, natural language processing (NLP), computational linguistics, robotics, and control theory.