Deep Learning and the Game of Go

I am a software engineer. The new AI paradigm is very important and this book is really very good at explaining this seminal work: AlphaGo.The examples are in Python however the text is sufficient to understand the importance parts of the algorithm.Highly recommended for software engineers.

I've read most ML books out there and none are as informative as this one.It'll teach you the basics of ML, how you can encode features and outputs using DNN is invaluable and an important topic which is often missed in most books. Go is used as an excellent case study and the historical design decisions are well documented and give you an amazing insider look of how to think about solving new ML problems.So much better than some garbage where you predict rent prices or how many hot dog carts new york needs.

Overall, I thought this book was pretty good. I've read a bunch of the O'Reilly books, but I thought this one approached the topic in a better way, for the most part.On the one hand, it makes a lot of assumptions, primarily about your coding ability, familiarity with machine learning and mathematics generally.On the other hand, it doesn't really get bogged down with tangents. The authors did a good job focusing on the task at hand, and methodically go through building up a deep learning system that comes together to play Go, while still touching on the things you need to know.Ultimately, I thought this book was pretty good, and I recommend it.My one real complaint is that -- across 350 pages -- it still felt like there wasn't a ton of depth to things. That's not necessarily terrible; it seems like their goal was to get the reader to a place where they could assemble their own deep learning Go-bot, rather than exhaustively cover all of deep learning. But it also seems like their intended audience was people who are very new to deep learning, so it kind of felt like I was learning just enough to accomplish the task at hand.

Learn by doing. This might be my favorite book I’ve ever read on ML.

The code won't work with the latest Tensorflow 2. The online demo has some serious bug which has been reported on Github but has not been fixed yet.If you following along with each branch on Github for different chapters as the authors suggested, you probably won't make it run successfully without doing a series of troubleshooting.An example is in chapter 4, the source code on Github (branch chapter_4) is missing one key module in agent "naive_fast.py", which I have to go to another branch to get it, which makes me question whether they've tested the code before publish it.Another example is if you just follow along the master brach on Github, you'll quickly get lost as there are lot's of additional code which are not explained in the book, at least they should write some useful comments to explain the new code.Speaking of comments, as other users also pointed out, the source code could really write some better comments.Finally, personally I think chapter 8 is better put at the appendix, which just spend a lot of time trying to explain how to deploy the UNFINISHED PROJECT to AWS cloud and compete against other online Go bots. I'd really appreciate if we can finished the entire project first, but that's just my own perspective.I've spotted some typos in the book, especially in some of the code snippets.

This book does a good walkthrough of the basics of machine learning for establishing the go game front end, plus standards used for exchanging Go information, both overall game and move by move basis. Then it adds how to set up a deep learning system. It also gos into some of the basics of playing go, and some of the variations as well.

From the basic rules of Go to the deep reinforcement learning, this book covers everything you should know about developing an AlphaGo-like game playing AI. Example codes are explained clearly and you can build a Go AI just following the instructions in the book.

It started out good, albeit lacking in code commentary. Starting at chapter 4, which has pretty critical components to understand, it becomes difficult to follow with no indication of where the code should go in terms of file structure. From what I can tell, the chapter has no mention of any file in which and of the code snippets should go. There's no perspective on how the code in the chapter fits at a higher level.Sure I could look at the code in Github, but I got the book to provide context and incremental understanding. The code on Github isn't well commented either which makes it difficult to understand which parts were written when in terms of chapter progression.Overall, there are a lot of moving parts and concepts to grasp and the lack of clarity makes it very frustrating. I'll keep going and see if it gets any better and update.