I've been interested in machine learning for a while. I started learning on my own, through online resources, and by trying to implement algorithms that seemed interesting. Soon this wasn't enough. Just seeing algorithms work is great, but I NEEDED more, I needed to understand why and how, or a at least the basic logic behind. Where did I turn? MOOC's and textbooks.
I enjoyed taking MOOC's, they gave me nice, followable overviews of general topics. The problem was, they were often too general. Textbooks came to the rescue when I once again required a deeper view.
This here is a short, but genuine 'roadmap' as the website names it.
http://metacademy.org/roadmaps/cjrd/level-up-your-ml
I recommend anyone interested in learning more about Machine Learning to start by picking up a textbook. The listed books provide a nice place to get started with ML depending on your existing level of knowledge. Even if not from this list, pick a textbook and stick with it. You can learn so much by soaking in the knowledge down to the spine of the textbook. Turn to other books as you need deeper understanding of specific topics.
Me?
I'm currently at a journeyman level in this list. The textbook I've been going through over the last month was Pattern Recognition and Machine Learning, and turns out the list would have given me the same book :D I will try the capstone project after getting a little more comfortable with the textbook contents, and have a little time to play around with what I learned.
Learn and play and learn more... No matter what the textbook is, thats a journey I'd like to continue.
