Let’s begin this blog the correct way, with a confession. As an undergrad math major, I did not take any statistics. There was a required course, called something like Mathematical Statistics. But my senior year arrived and I had not taken it. Instead of biting the bullet, I spoke with my faculty advisor and he arranged with the department to waive the requirement. It is one of my regrets.
Despite my confession I have picked up basic ideas about statistics over the years. In graduate school/post-doc years I had enough conversations (and saw enough talks) that I can tell you about i.i.d. random variables, covariance, cdf’s and pdf’s. But statistical ideas and considerations never became as natural to me as algebraic and geometric/topological ones did.
Well, the day of reckoning has come. I want to barge my way into the world of mathematical applications — machine learning, particularly. And statistics will be front and center. So please, do bear with me if you see my statistical gaps. I am learning too.
In the blog, I will be drawing on a few sources. Mainly, this includes the following references.
- Understanding Machine Learning: from theory to algorithms by S. Shalev-Shwartz and S. Ben-David.
- Mathematics of Machine Learning lecture notes from the MIT OCW course by Philippe Rigollet.