I don't know anything about this book, but in general the pre-reqs for any form of ML (unless you are working in the abstract or with a "black-box" library or s

I don't know anything about this book, but in general the pre-reqs for any form of ML (unless you are working in the abstract or with a "black-box" library or such) are a good working knowledge in statistics/probabilities, and an understanding of linear algebra (mainly how to manipulate and work with vectors/matrices). Oh - and an understanding of when and how to "parallelize" problems can be important, too. ML and associated problems aren't the most intuitive to understand, solve, etc - but they are a great challenge to expose yourself to if you have the interest. /disclosure: I'm currently enrolled in and participating in Udacity's Self-Driving Car Engineer nanodegree, plus I have participated in and completed their CS373 program, and I was a part of (and completed) the original Stanford ML Class MOOC that was taught by Andrew Ng. I guess you can say I am a bit biased on this subject...?