Deep Learning (color)

Deep learning is a type of machine learning that empowers computers to learn from experience and understand the world in terms of a hierarchy of concepts. Since the computer acquires knowledge from experience, there is no need for a human operator who formally describes the knowledge necessary for the computer. Hierarchical organization allows a computer to learn complex concepts by constructing them from simpler ones; the graph of such a hierarchy can contain many levels. In this book, the reader will find a broad overview of the topics studied in deep learning. The book contains mathematical and conceptual foundations of linear algebra, probability theory and information theory, numerical calculations and machine learning to the extent necessary to understand the material. Deep learning techniques used in practice are described, including deep direct distribution networks, regularization, optimization algorithms, convolutional networks, sequence modeling, etc. Applications such as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics and video games are considered. Finally, promising areas of research are described: linear factor models, auto-encoders, representation learning, structural probabilistic models, Monte Carlo methods, statistical sum, approximate inference and deep generating models. The publication will be useful for students and postgraduates, as well as experienced programmers who would like to apply deep learning as part of their products or platforms.