Genetic Algorithms and Machine Learning for Programmers: Create AI Models and Evolve Solutions (Pragmatic Programmers)

Finally, someone who explains this so students can understand it. This is hands down the book you want to start with. The other popular book groking algorithms has a wonderful first chapter but the rest of the book is meh. This book is enlightening all the way through. Thank you for writing it!

Just reinforcing what others have said. The Kindle version's formulas are unreadable. From the text, it sounds like a nice discussion of entropy. Too bad it's incomprehensible. Surprised no one has responded or corrected the problem.

Formulas cannot be properly read in Kindle Fire tablet, cannot be properly read in Android Kindle App, and cannot be properly read in Kindle Cloud Reader.Do your job, both Amazon and the publisher, and get your act together.I was able to return the Kindle version for a refund. Have ordered print copy. When the print copy arrives, I will post an updated review. But be forewarned, the publisher didn't bother to get their Kindle formatting right.

I love this textbook and enjoying it so far!

I am a software developer by trade and I have always been interested specifically in artificial intelligence (and to a lesser degree, computer learning). This book is an overview of some of the subtopics in these fields, though it is not typical in the approach. Instead of addressing each in a theoretical, scholarly manner like a textbook, it instead provides a much more hands-on approach and leads the reader through a series of exercises designed to demonstrate the material. The approach is engaging and actually fun.First, a word about computer languages: the author explains that the primary languages in used here are Python, C++ and Javascript, though other languages can be used. I personally like Perl and C++, so as I work through the examples and learn I have been using Perl, and it works well enough for me. I could possibly make life easier on myself by using a different language, but the idea is to learn the material, not struggle with a computer language.Also: It really helps to have a mathematical background. The author presents several mathematical formulas to back up the material and explain the methods, and for someone who isn’t mathematically inclined these will potentially be intimidating. If you aren’t strong in math this will hinder understanding but will not be a complete show-stopper.The author presents ten exercises that demonstrate an abstract topic in AI and/or machine learning. The first exercise is described as helping a turtle escape a bag. This is the same base process as a breadth-first search, though there are some additional factors in the overall system that complicate the calculations. The next exercise builds on the first and recreates the bag with data captured from the first exercise. Not all exercises are related, though, and many are independent of the others. One particular exercise provides an interesting approach to creating artificial life using the classic Game of Life as a starting point to a more complex variation.The formula this book follows is consistent across all of the exercises. The author presents the problem to be solved, a mission statement, a how-to section, implementation, analysis of the result and finally topics for further learning. This works well.This is not a comprehensive study on either machine learning or artificial intelligence. If it was so, it would be a huge book weighing several pounds (and probably would have more errors due to the sheer quantity of material). What is here, though, is a good starting point with a fun introduction. As I worked with this book I felt it would be a good book for use in a classroom as a lab book for advanced students.If you have an interest in computer learning or artificial intelligence, this is a nice starting point. It’s not as heady and intellectual as many others are and it does not rely on the theory to carry itself. Instead, it is a work book that can be done outside of a classroom and worked at your own pace. The material is interesting and the exercises more engaging. It isn’t a comprehensive study of the whole field, but more of an involved introduction. This is an easy book to recommend, based on my work with it.

This one is hard. My degree is in computer science. My master's thesis was on maching learning. I still had to read parts of this twice to understand it completely. That said, I think this is one of the most comprehensive texts on the subject I've seen in a long time. It covers not just how the algorithms work, but why and how to train and tweak them. I know neural nets are the current darling of AI right now, but I've always loved the GA, and it's nice to see such an excellent coverage of it.

The product is mostly useless the way it is currently being sold. I just threw money away.

I received the print edition of this book for review. So for some background, I've been a programmer for over 30 years. I did research in GA's in the late 90s, and have not kept up with GA/ML/NN since then. I have gone through quite a few of the exercises and found the analogies and instruction to be excellent.This is exactly the book I needed to get oriented again. I found the brief histories of the technologies accurate and interesting, and the author leaves lots of little tidbits for you to expand your knowledge, such as the discussion of L-systems (which I honestly have not thought about since the late 80s).If you're at all interested in the subject matter and need an introductory text, have at it. This is a fantastic book.

It's a lovely book which guides you carefully through the different level of deep learning. The examples are inspired by real-world problems and help a lot for visibility. The coded examples of the webpage are a little nasty and should be reformated.

An excellent introduction to the subject, accessible and clearly written but covering a good deal of ground and encouraging the reader to build and experiment with a range of machine learning techniques.