OK, so here I am, a regular guy with alot of wasted time in the past. Not especially bright, with a modest working memory capacity and a lacking self control. Not that young either. Should that stop me from learning about machine/statistical learning? No fucking way, I don't give up that easily.
So here's what I did... I grabbed a book called "An Introduction to Statistical Learning", written by Hastie et. al. It's definitely not math-heavy (actually there's almost no math), which I suppose is not that great, but I'm thinking it's not that bad for an intro book. Was that a good call?
I do know a little bit of probability theory, statistics, matrix algebra, graph theory, and some basic optimization principles (I'm not familiar with proofs, just basic concepts; I understand what is overfitting/underfitting, what's gradient descent, bias-variance trade-off, conditional probability, Dijkstra's algorithm etc.).
I am able to code a thing or two in Python and I'm trying to learn R. Should I stick to that?
Maybe I should put it this way - what is the shortest path from this point in time, to the one where I'm able to say that I consider myself to be data scientist/machine learning expert? What would be my optimal strategy (topics to learn about) and my optimal tactic (books, online courses...)? I don't want to learn it quickly, I want to learn it thoroughly. Just don't want to waste my time on things that won't be of any use.
Dealing with data makes me happy so learning won't be a problem, but if you guys could teach me how to learn, I'd appreciate it alot.
[link][8 comments]