Artificial intelligence. A modern approach. Volume 3. Learning, Perception and Action

The most complete and up-to-date introduction to the theory and practice of artificial intelligence! In the fourth, updated, revised and expanded edition of this book, the field of artificial intelligence (AI) is explored and analyzed in all its vastness and depth. All modern achievements are presented here and ideas that have been formulated in research conducted over the past fifty years are presented, as well as collected over two millennia in areas of knowledge that have become an incentive for the development of AI as a science. Previous editions of this book have become classic examples of AI literature and have been accepted as a textbook in more than 1,400 universities in 128 countries around the world, where they were highly appreciated as a convincing summary of the results achieved in this field of science. The book is supplemented with an extensive set of Internet resources, including exercises, software and research projects, algorithm implementations, as well as additional materials and links for students and teachers. What's new in the fourth edition In the fourth edition, the reader will get acquainted with the latest technologies and concepts presented in a more unified form with a new or expanded coverage of topics such as machine learning, deep learning, transfer learning, multi-agent systems, robotics, natural language processing, the problem of causality, probabilistic programming, as well as confidentiality, impartiality and AI security. What is presented in this volume Due to the exceptionally large volume of this encyclopedic edition, its Russian-language version is presented in the form of a three-volume book. This third volume, subtitled Learning, Perception and Action, examines the most complex and modern subdomains of AI: various approaches to machine learning, natural language processing, computer vision and robotics, as well as equally important aspects of AI - its confidentiality, impartiality and security. That in other volumes The two previous volumes I and II dealt with the basics of AI theory: intelligent agents and a possible approach to problem solving (knowledge, reasoning and planning (Volume I)) and the action of agents in an environment more close to the real (knowledge, reasoning and decision-making under conditions of uncertainty, including in a multi-agent environment (volume II)). 4th edition.