Jordan Ellenberg is the John D. MacArthur Professor of Mathematics and a Vilas Distinguished Achievement Professor of Mathematics. After earning his doctorate from Harvard University in 1998 and teaching for several years at Princeton University, he came to the UW in 2005. A two-time gold-medalist in the International Mathematical Olympiad, he is now a Discovery Fellow at the Wisconsin Institute for Discovery, where he works with the Machine Learning group and the Institute for Foundations of Data Science. He’s also the author of How Not to Be Wrong, published in 2014, and Shape, published in 2021. (He was also recently featured in a Jeopardy! clue.)
Chief Area of Research:
I’m a mathematician — mostly a pure mathematician. And my main areas are number theory and algebraic geometry. I think to most people, if you say, “Oh, I work in the kind of math that has to do with numbers,” people would say, “Well, isn’t that all of it?” And that’s totally not true. In our department, people work on all kinds of things, ranging from how to mathematically model the weather, to deep questions about probability, to questions about functions. So, number theory is kind of like this very classical part of math that deals with the kinds of questions that people mostly think math is about, like equations and solutions.
On The UW Now, I’ll Discuss:
We’re supposed to talk about predictions, and I don’t make predictions. But I’ll talk about the limitations of predictions and what we can expect, and maybe how you assess whether a prediction is good or not after the fact.
One Thing I’d Like Viewers to Remember Is:
A prediction isn’t bad just because it was wrong. And it wasn’t good just because it was right.
A prediction can be a good prediction even if it doesn’t come true. If the worst hitter in the league comes up against the best pitcher in the league, a good prediction is probably going to be that the guy’s going to strike out. And even if he hits a home run, it doesn’t mean your prediction was a bad prediction. So, I think that’s something that’s important to keep in mind as we retroactively think about this game of prediction.
To Get Smart Fast, Read:
There’s a nice discussion of calibration that Nate Silver does.