I'm a math geek, but I'm also a mostly self-taught data scientist.<p>"The Elements of Statistical Learning" (<a href="https://web.stanford.edu/~hastie/local.ftp/Springer/OLD/ESLII_print4.pdf" rel="nofollow">https://web.stanford.edu/~hastie/local.ftp/Springer/OLD/ESLI...</a>) is far and away the best book I've seen.<p>It took me hundreds of hours to get through it, but if you're looking to understand things at a pretty deep level, I'd say it's well-worth it.<p>Even if you stop at chapter 3, you'll still know more than most people, and you'll have a great foundation.<p>Hope this helps!
Just a curiosity: One of the authors also proposed Pegasos SVM [1] which is a nice online approximation to SVM and that can be written in 15 lines of code or so.<p><a href="http://ttic.uchicago.edu/~nati/Publications/PegasosMPB.pdf" rel="nofollow">http://ttic.uchicago.edu/~nati/Publications/PegasosMPB.pdf</a>
I feel like the barrier to machine learning for me, as I've seen in many tutorials and books and is an immediate discouragement, is the massive amount of math thrown in your face. Many of us didn't just graduate, need glasses and fall asleep at 8pm on the couch when the kids are in bed... Math is this distant fragment of memory buried under years of everything not Math.<p>It feels like machine learning is only taught by academia but the majority of the audience is for practical use by the average developer wanting to play with it today.
I've read this book and warmly recommend it. It has a very pragmatic "no bullshit" approach and it's very mathematical and concise.<p>The neural networks chapter is tiny (but that's ok - that's not the focus) and some of the questions are really hard - but overall I've really enjoyed it.
I am considering taking Udacity's machine learning nanodegree [0] with zero machine learning background. It seems interesting. Any thoughts?<p>[0] <a href="https://www.udacity.com/course/machine-learning-engineer-nanodegree--nd009" rel="nofollow">https://www.udacity.com/course/machine-learning-engineer-nan...</a>
It's pretty cool that the book is not only free, but they link to courses that uses it.<p>If you speak Hebrew you can get two different professors take on how to teach the material in the book, as well as lecture notes from a total of 3 professors. That's pretty neat if there's a concept you are struggling with as a student!
Glad to see another book to learn from for free. But my problem now is that there are so many books each with somewhat different approach and content for the same ML techniques. Not necessarily bad, but I get somewhat confused when trying to apply a method.<p>EDIT: I guess the focus on the theory might help me.
My question for someone that has an intermediate level of skill in machine learning, what's the best way to dip your toes in? (Udacity, coursera, edx, PDFs, Talking Machines podcast, etc)
There's also the freely-accessible book <i>A Course in Machine Learning</i>:<p><a href="http://ciml.info/" rel="nofollow">http://ciml.info/</a>
Related: Foundations of Data Science: <a href="http://www.cs.cornell.edu/jeh/nosolutions90413.pdf" rel="nofollow">http://www.cs.cornell.edu/jeh/nosolutions90413.pdf</a>
Took the course at the Hebrew University. Awesome course. Here are the slides (English) and videos(Hebrew): <a href="http://www.cs.huji.ac.il/~shais/IML2014.html" rel="nofollow">http://www.cs.huji.ac.il/~shais/IML2014.html</a>
Top comment in another thread whines that nobody understands what the machinery of ML algorithms are really doing.<p>Top comment here whines that math is hard.