Artificial Intelligence: A Modern Approach. The Definitive AI Textbook!

The following is a review of the book Artificial Intelligence: A Modern Approach (Pearson Series in Artifical Intelligence) by Stuart Russell and Peter Norvig.

Review of Artificial Intelligence: A Modern Approach

Artificial Intelligence is a big field, and this is a big book. Covering ALL of the different topics on the field of Artificial Intelligence in a way that is practical, highlights the applications, difficulties, theories, and mathematics, could only be done in such a tremendous piece of work.

Firstly published in 1995, Artificial Intelligence: A Modern approach has been regularly updated since then, reaching this fantastic fourth edition in 2020. The book comes in a hard-cover paperback format that stands out for its quality and finish.

This book has been called “The most popular Artificial Intelligence text book in the world“. Let us explain why.

Since its first publication, it has been widely used in universities across the world in courses that range from Basic Artificial Intelligence, to Statistical Reasoning or Robotics.

It presents an incredibly wide range of concepts related to Artificial Intelligence in an unified manner, and offers coverage of machine learning, deep learning, transfer learning, multi-agent systems, robotics, natural language processing, causality, probabilistic programming, privacy, fairness, and safe AI, amongst many other topics.

It explores the full breadth of the field, which encompasses logic, probability, and continuous mathematics; perception, reasoning, learning, and action; fairness, trust, social good, and safety; and applications that range from microelectronic devices to robotic planetary explorers.

All of this while uniting the ideas that have emerged over the past decades in the field of AI, with statistics and mathematics, while avoiding excessive nonsense and formalities.

Mathematical formulas and Pseudocode are included, but the book is neither for mathematicians nor programmers. Rather, it is oriented for a technical university student audience, that wants to go deep and understand all the core concepts surrounding AI.

Overall, it is a very pleasant read, it is easy to follow, and the writers build up complex AI concepts piece-by-piece, breaking them down into easy, understandable bits. Finally, the examples are contemporary, applied, and up to date with the newest advances. An amazing read!

It is an amazing book to learn in an structured manner about the world of AI and all that surrounds it, however, if what you are looking for is getting dirty and building Machine Learning or Artificial Intelligence Applications, you should probably check out other texts like Building Machine Learning Powered Applications, Python Machine Learning, or Hands-On Machine Learning with Scikit-Learn & TensorFlow.

Contents of AI – A Modern Approach

The book has 28 chapters that are divided into 7 parts. It is incredibly long, however it is not a book that we would sit down to read in one go, but more of a manual to learn specific topics through great, delightful explanations.

• Part I: Artificial Intelligence – Sets the stage for the following sections by viewing AI systems as intelligent agents that can decide what actions to take and when to take them.
• Part II: Problem-solving – Focuses on methods for deciding what action to take when needing to think several steps ahead such as playing a game of chess.
• Part III: Knowledge and reasoning – Discusses ways to represent knowledge about the intelligent agents’ environment and how to reason logically with that knowledge.
• Part IV: Uncertain knowledge and reasoning – This section is analogous to Parts III, but deals with reasoning and decision-making in the presence of uncertainty in the environment.
• Part V: Learning – Describes ways for generating knowledge required by the decision-making components and introduces a new component: the Artificial Neural Network
• Part VI: Communicating, perceiving, and acting – Concentrates on ways an intelligent agent can perceive its environment whether by touch or vision.
• Part VII: Conclusions – Considers the past and future of AI by discussing what AI really is and why it has succeeded to some degree. Also discusses the views of those philosophers who believe that AI can never succeed.

Programs in the book are presented in Pseudo-code, with materials in Java and Python available online. These materials can be found in the following Github Repo.

In the following link you can also find the official website of the book to deeply dig into what it contains: Official Pearson Publishing entry.

Lastly, some chapters have contributions of guest writters, Like Ian Goodfellow, author of Deep Learning in the Deep Neural Networks chapter, or Anca Dragan on the robotics, chapter.

Highlights of this last edition include stellar contributing guest writers. For example, Ian Goodfellow contributed to the chapter on deep learning and Anca Dragan contributed to the robotics chapter.

Summary of Artificial Intelligence: A Modern Approach

This is the most comprehensive book on the field of Artificial Intelligence available today.

While it might not be the best AI book to start building Artificial Intelligence projects or programming your own little applications, that is not it is goal.

In its 1,000 + pages, it will give you a broad and fluid coverage of the diverse field of AI.

You will get an Introduction to Artificial Intelligence from all its different angles, learn about the future of AI, the ethics implied, and some awesome applications like the fusion between robotics and Artificial Intelligence.

It is the most recommended textbook on AI in education for a reason. You can find it at the best price on Amazon here:

Artificial Intelligence: A Modern Approach

Thank you for reading, we hope you liked the review of Artificial Intelligence: A Modern Approach, and have a great day!

Tags: Artificial Intelligence Book, Artificial Intelligence a Modern approach, AI book, AI Modern approach, Artificial Intelligence.