Mastering Reinforcement Learning with Python: Build next-generation, self-learning models using reinforcement learning techniques and best practices

The flow of the book is great, it is easy to follow the book and the python codes concurrently. I strongly recommend the book everyone including the ones with no strong background in machine/reinforcement learning.

Must have if you are into applied RL.

Great RL introduction. I've been going through this book for the past couple of months. I appreciate the code snippets that are present. I like that the paper references some lectures from Berkeley which I can check out online.

The book provides good level of details on foundation of RL, practical tools and libraries, and step by step guides on solving some applied problems. You need to have some knowledge in statistics and probability to understand the topics discussed in the book. Some programming skills in python is also required but you do not need to be an advanced python programmer to benefit from the book.There are links to many useful external resources and blog posts to help you gain deeper knowledge in topics discussed at each chapter. Many advanced topics such as Machine teaching is covered that has industrial relevance (example: Microsoft's project bonsai). Most importantly, the challenges of applying RL and the limitations of RL for some applications are discussed. Being aware of the limitations can save you a lot of time in your solution formulations.

This is a book for practitioners. It is well written and covers a wide range of topics from the basics of RL and Markov decision processes to multi-agent systems. It focuses on modern methods of deep RL including model-based approaches, notably also an introduction to machine teaching. Very nice is also part 4, with a lot of application examples from robotics, supply chain management, marketing and cybersecurity. I definitely recommend it for everyone interested in developing their own real-world RL solutions.

It’s a very good reference book for beginners and experienced engineers.

First thing first, the book is not for beginners. If you are a beginner, you want to start with some ddqn tutorials.If you are an intermediate reinforcement learning practitioner, this book is for you to get your stuff deployed.You won't want to cover-to-cover-read the book. Treat it like reference material.If you want to get the most out of this book, give the full index a skim, and find the thing you wanted to deploy/learn about. Read that, then the book will tell you how to do it in Ray-rllib. If you are already familiar with rllib, great. If you aren't, skim to middlish introductory portions of the book where rllib is first introduced. In case you forget, there's a bunch of reference explanations of the algorithms in the first 200 pages.For programming reference books, this is on the money. If you know what to expect when you get the book, you wont be disappointed.

I am a few chapters in and really digging that this is a truly a book with hands-on examples. Would love for it to go just a little more in depth on the “why” things are happening or decisions are made, but all up has been a really fun and impactful read.