compneuroer

I went through this exact same decision-making process, and am currently doing a phd in a subfield of ML. I'm happy with it because I realized I would actually be bored as a doctor, and especially in med school--it's just not the type of intellectual work I'd like to be doing. A PhD in stats gives you a lot more flexibility than you would expect, and you don't have to work in your exact phd discipline later on. You also don't have to be a Yann Lecun to make contributions to a field. Just some other things to consider!

It's a good question -- it may be a bit late for this now, but if you have specific professors in mind that you want to work with, maybe you can narrow your list of colleges down based on those professors? A masters and 1 years research experience is pretty solid but not a crazy amount.