March 2, 2017

Getting up to speed on stats and machine learning

Two students, two great guides to understanding complex topics

One of the things I knew I wanted to do when I came to Harvard this year for the Nieman Fellowship is get a better handle on a few topics I often struggle to understand and apply, namely statistics and topics in artificial intelligence.

I took courses in both last semester. But it helps to have good reference guides, especially when they're clear and concise. So one lesson I've learned this week: When it comes to creating crib sheets, never underestimate the abilities of undergraduates.

Over at Brown University, senior Daniel Kunin created a site called Seeing Theory, a beautiful visual introduction to probability and statistics. The site includes the exploration of topics from basic probability to linear regression (all using the D3 library). The visualizations are also interactive, which really allows you to get a sense of how the theories of statistics apply in practice.

Simulating a dice roll, 100 rolls at a time. // Graphic by Daniel Kunin

Here at Harvard, senior Noah Yonack wrote up a great primer on machine learning for those of us without an extensive background in computer science and mathematics. It's 13 pages long, but does a great job introducing the basic vocabulary and techniques.

Right up front, it gets at a really fundamental question I had: What's the difference between machine learning and artificial intelligence?

 

The full document is below, shared with permission from Noah (thanks Noah!).

A Non-Technical Introduction to Machine Learning

by Noah Yonack