Theory Day at Georgia Tech (and also in NYC)

2011-11-28 by . 0 comments

Post to Twitter

This post is by David Pritchard and Lev Reyzin.

Last Friday (11/11) was Theory Day at Georgia Tech. In fact, it was also Theory Day in NYC, although nobody seems to think the choice of date was coordinated or the start of a new national holiday. In Atlanta, Georgia Tech invited Avi Wigderson to give two talks on Thursday (making for two theory days in a row!), one for the broader public and then a math-on-whiteboard talk in the afternoon; and on Friday, there were four talks: by Thomas Dueholm Hansen, Mohit Singh, Alexander Mądry, and Ryan Williams, all aimed at theoreticians.

Dave: I was pretty impressed by the quality of the talks, which were about an hour long, and therefore gave the speakers quite a good chance to get into a bit of the technical details that are usually skipped at conferences. I have skimmed a couple of the papers before and really got much more intuition with these in-person explanations. Avi’s second talk was probably the most self-contained, on aspects of coding and a generalized Erdős/Gallai/Melchior/Sylvester theorem.  Also, we found that Alex may be a pyromaniac, since his examples for the k-server problem involved houses burning down (which you could think of as data requests). I was also pretty impressed by the quality of the lunch, which was a buffet of Indian food.

Lev: While I was a graduate student at Yale and during my year at Yahoo! Research in New York, I attended all the New York Theory Days for 5 years straight, so I was glad to see catch on in Atlanta too.  Apparently, we have Zvi Galil (now Georgia Tech’s College of Computing dean) to thank for the ideas of starting both the New York and the Atlanta theory days — so while the dates were not coordinated, the Theory Days, in some sense, were.  All four talks were great.  I especially enjoyed Alexander Mądry’s great and accessible talk on his recent progress on the k-server conjecture.  Ryan Williams also gave a very nice talk on how algorithms for circuits can imply lower bounds; it gave me some intuitive understanding that I didn’t have before.  Finally, I should note that I seem to remember the talks being filmed, so I’m hoping the videos will show on on Georgia Tech’s ARC website sometime.

If any readers attended the NYC Theory Day, please share your impressions in the comments!

Subscribe to comments with RSS.

Leave a comment

Log in
with Stack Exchange
or