Ranked in these QuestionsQuestion Ranking
Pro Has good balance between technical and non-technical information
Written to cover both the subject matter and theory.
Pro Covers 3 essential reinforcement learning approaches
The second part of the book covers Dynamic Programming, Monte Carlo Methods and Temporal Difference Learning learning methods. They're presented in an interesting way and explained through various examples.
Available online for free.
Pro Offers lots of real world examples
Con Uses vague terms
Supplementary resources that have formalized the terms used in this book might be needed to fully understand everything.