[arbitrage314]

I'm a math geek, but I'm also a mostly self-taught data scientist.

"The Elements of Statistical Learning" (https://web.stanford.edu/~hastie/local.ftp/Springer/OLD/ESLI...) is far and away the best book I've seen.

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.

Even if you stop at chapter 3, you'll still know more than most people, and you'll have a great foundation.

Hope this helps!

[reader5000]

Having read significant chunks of both ESL and Understanding Machine Learning (albeit UML much more recently) I would argue that for many readers UML is superior.

ESL pays short shrift to the computational complexity of learning whereas UML explicitly handles both statistical and computational complexity concerns. It doesnt matter how statistically pure your algorithm is if its running time scales exponentially with your data.

All of UML's chapters are conceptually unified even when discussing different ML algorithms, with ESL being more of a grab-bag by chapter.

Still, both high quality and free!

[arbitrage314]

Thanks for the interesting comparison!

[gajjanag]

I am a graduate student at MIT, and can second this recommendation. It is a fantastic book for machine learning and nothing else I have seen comes close.

[nafizh]

You meant ESL or UML?

[lakeeffect]

ESL. His post was an hour before the reader5000 uml post.

[Thriptic]

Would you say elements is superior to Bishop's book?