What do you think?
Rate this book
Deep Learning and the Game of Go
Summary
Deep Learning and the Game of Go teaches you how to apply the power of deep learning to complex reasoning tasks by building a Go-playing AI. After exposing you to the foundations of machine and deep learning, you'll use Python to build a bot and then teach it the rules of the game.
Foreword by Thore Graepel, DeepMind
Purchase of the print book includes a free eBook in PDF, Kindle, and ePub formats from Manning Publications.
About the Technology
The ancient strategy game of Go is an incredible case study for AI. In 2016, a deep learning-based system shocked the Go world by defeating a world champion. Shortly after that, the upgraded AlphaGo Zero crushed the original bot by using deep reinforcement learning to master the game. Now, you can learn those same deep learning techniques by building your own Go bot!
About the Book
Deep Learning and the Game of Go introduces deep learning by teaching you to build a Go-winning bot. As you progress, you'll apply increasingly complex training techniques and strategies using the Python deep learning library Keras. You'll enjoy watching your bot master the game of Go, and along the way, you'll discover how to apply your new deep learning skills to a wide range of other scenarios!
What's inside
About the Reader
All you need are basic Python skills and high school-level math. No deep learning experience required.
About the Author
Max Pumperla and Kevin Ferguson are experienced deep learning specialists skilled in distributed systems and data science. Together, Max and Kevin built the open source bot BetaGo.
Table of Contents
Deep Learning and the Game of Go teaches you how to apply the power of deep learning to complex reasoning tasks by building a Go-playing AI. After exposing you to the foundations of machine and deep learning, you'll use Python to build a bot and then teach it the rules of the game.
Foreword by Thore Graepel, DeepMind
Purchase of the print book includes a free eBook in PDF, Kindle, and ePub formats from Manning Publications.
About the Technology
The ancient strategy game of Go is an incredible case study for AI. In 2016, a deep learning-based system shocked the Go world by defeating a world champion. Shortly after that, the upgraded AlphaGo Zero crushed the original bot by using deep reinforcement learning to master the game. Now, you can learn those same deep learning techniques by building your own Go bot!
About the Book
Deep Learning and the Game of Go introduces deep learning by teaching you to build a Go-winning bot. As you progress, you'll apply increasingly complex training techniques and strategies using the Python deep learning library Keras. You'll enjoy watching your bot master the game of Go, and along the way, you'll discover how to apply your new deep learning skills to a wide range of other scenarios!
What's inside
About the Reader
All you need are basic Python skills and high school-level math. No deep learning experience required.
About the Author
Max Pumperla and Kevin Ferguson are experienced deep learning specialists skilled in distributed systems and data science. Together, Max and Kevin built the open source bot BetaGo.
Table of Contents
380 pages, Paperback
Published January 25, 2019
24 people are currently reading
124 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 6 of 6 reviews
August 18, 2018
Good hands-on tutorial. However, the reinforcement learning part (last two chapters) hides the forest of the whole RL field by only showing one tree.
Meanwhile, the code repo seems focus on the Actor Critic developed by deepmind. The AC is said to beat DQN by large margin, however, the DQN is not mentioned anywhere in the book.
Meanwhile, the code repo seems focus on the Actor Critic developed by deepmind. The AC is said to beat DQN by large margin, however, the DQN is not mentioned anywhere in the book.
March 28, 2019
an enjoyable guide to an area I knew nothing about. I will now try put the lessons into practise
January 1, 2019
It covers many topics in some detail. Many of the complexities of each subject are simplified.
July 17, 2020
The best book I have ever read about ML.
The thing about ML is that every individual idea is simple and relatively easy to understand and implement with Keras or whatever, but stitching the whole thing together into a working project that solves a real problem is where things become devilishly complicated. Books like Goodfellow's bible are good for understanding the principles and the math, and the OReilly sklearn book has snippets for all sorts of scenarios, but other than this book, I know of no other book that takes a full AI problem start to finish and walks you through all of the concepts and the code needed to solve the individual components and to put them all together into a working whole.
It almost seems as if Go is like the Hydrogen atom of AI--a paradigmatic example simple enough that it can be fully treated in a single book-length study, yet complex enough that if you can master it, you've acquired a true sense of how deep learning actually works.
I've gone through this book in considerable detail with a number of colleagues at the post-doctoral level, as well as with graduate students, and we have all learned greatly from it. If you are interested in taking your knowledge of cat detectors and Towards Data Science tutorials and transforming yourself into an AI/ML problem solver, I can think of no better single book than this.
The thing about ML is that every individual idea is simple and relatively easy to understand and implement with Keras or whatever, but stitching the whole thing together into a working project that solves a real problem is where things become devilishly complicated. Books like Goodfellow's bible are good for understanding the principles and the math, and the OReilly sklearn book has snippets for all sorts of scenarios, but other than this book, I know of no other book that takes a full AI problem start to finish and walks you through all of the concepts and the code needed to solve the individual components and to put them all together into a working whole.
It almost seems as if Go is like the Hydrogen atom of AI--a paradigmatic example simple enough that it can be fully treated in a single book-length study, yet complex enough that if you can master it, you've acquired a true sense of how deep learning actually works.
I've gone through this book in considerable detail with a number of colleagues at the post-doctoral level, as well as with graduate students, and we have all learned greatly from it. If you are interested in taking your knowledge of cat detectors and Towards Data Science tutorials and transforming yourself into an AI/ML problem solver, I can think of no better single book than this.
May 30, 2021
I reckon it's an unenviable task to write a book about sophisticated machine learning bots aimed for working engineers -- and I'm not sure Pumperla and Ferguson were exactly up to the task either. I don't know how useful this is as something to just plunder out whatever you need for your own project, and I'd be hesitant to attach any curriculum to this.
Data science no doubt requires experimenting yourself to build intuition, but the authors seem to give up on giving any more than the bare minimum or surface level explanation (there's lot of diagrams and code snippets, but they're honestly noise in how they're presented). The question "But why?" came up a lot, and the answer I got more often than not was "Because this works better." I know that they know why, but I'm left to dig out what I need on my own. Which is fine, I guess, as a pedagogy...but it makes you wonder what's the point of this as a book. Honestly makes more sense to just grab the code from their github, play with it myself and some articles online, and not bother with the book at all.
It's still very interesting, and I do feel I came out with some better appreciation of tech behind Alpha Go/Alpha Go Zero, but I think there's some better resources in the Manning book line or online.
Data science no doubt requires experimenting yourself to build intuition, but the authors seem to give up on giving any more than the bare minimum or surface level explanation (there's lot of diagrams and code snippets, but they're honestly noise in how they're presented). The question "But why?" came up a lot, and the answer I got more often than not was "Because this works better." I know that they know why, but I'm left to dig out what I need on my own. Which is fine, I guess, as a pedagogy...but it makes you wonder what's the point of this as a book. Honestly makes more sense to just grab the code from their github, play with it myself and some articles online, and not bother with the book at all.
It's still very interesting, and I do feel I came out with some better appreciation of tech behind Alpha Go/Alpha Go Zero, but I think there's some better resources in the Manning book line or online.
August 14, 2022
Initially I was blown away by this book; it was inspiring and informative, opening my eyes to all sorts of ways of thinking and new approaches to difficult problems. But after a bit of distance a lot of cracks have appeared. There are quite a few places where there is a bit of hand waving and you're told you can download the relevant code from their GitHub account, and then they blow past it and you're on to the next thing without a backward glance. All in all it covers a great deal and well, but don't let it fool you into thinking you're now a master of deep learning. Take it as a first stepping stone into this incredible new field.
Displaying 1 - 6 of 6 reviews