I've been watching some introduction Machine Learning videos just to see what it's all about and so far my takeaway has been that it just involves learning a bunch of algos (regression, k means, decision trees etc...) and how they work and when to apply them.

I've had an interest in statistics and am hoping to make a transition into a career that's more stats oriented. I think it would be really cool to eventually work on AI or something more research based (as opposed to just solving business problems), although not sure how feasible that is.

Is mastering Machine Learning really just learning about the different algos out there and knowing how/when to apply them? And what are you doing as a Machine Learning researcher? Are you trying to come up with more algos or just looking to do data analysis?

  • 1
    $\begingroup$ It's not only the how and when, but also why. Then, you transition to ML research, when you find yourself working on the why not. $\endgroup$ – Nutle Oct 1 '19 at 14:22
  • $\begingroup$ What else would it be? The machine has to learn somehow, and the way that machines do basically anything is via an algorithm. In fact, a lot of ML is more about algorithms that return algorithms. For example, simple linear regression is an algorithm than returns the parameters necessary to evaluate yet another algorithm that performs a prediction at a new point. $\endgroup$ – Scott Oct 1 '19 at 16:11
  • $\begingroup$ Ok thanks. I am just trying to figure out what to pay most attention to as I try to teach myself ML, given my goals. $\endgroup$ – confused Oct 1 '19 at 17:03