Update from a student (and an update on Erdos-Bacon numbers)

I got an e-mail from Casey Warmbrand, one of my former undergrad advisees and now a Ph.D. student in mathematics at University of Arizona.

He wrote to ask if I knew about Erdos-Bacon numbers. You might remember that I do, as evidenced by this blog entry from last year. The wikipedia entry for Erdos-Bacon numbers seems decently more extensive than it was last year. I especially like the section showing which actors have finite Erdos-Bacon numbers (e.g., Natalie Portman's is bounded above by 7).

One of my frivolous goals is to achieve a finite Erdos-Bacon number. Right now, my Erdos number is bounded above by 4 (and I have numerous paths of length 4). My Bacon number is infinite, but there is a small chance that that will eventually change. In particular, it's conceivable that I'll be interviewed for a documentary because I co-wrote/co-edited Legends of Caltech III, and my plan is to request that Kevin Bacon have a cameo in that flick in order to give me a Bacon number of 1 and an Erdos-Bacon number of 5 (which would be one of the lowest ones out there). When I had the chance, perhaps I should have asked if I could have a cameo in Starship Dave, the movie for which I briefly served as a mathematical consultant. (Be sure to check out the 'DEI' I inserted into the unified field theory that appears in that film...)

Before I go, I also want to give an update on what Casey is doing: He was a coauthor on my PNAS paper on Congressional committee networks. He decided he prefers pure math and is working for Ken McLaughlin at University of Arizona. (McLaughlin is quite famous for his work on integrable systems.) Casey successfully wrote and defended his Masters thesis, which concerned partitions given by the Plancherel measure and an asymptotic analysis analogous that of random matrix theory and the Wigner semicircle law. Casey's doctoral thesis will focus on domino tilings of the aztec diamond (and possibly tilings of a hexagon with
rhombi) and the use of orthogonal polynomials to relate the tilings (and the non-intersecting paths that describe them) to probability distributions related to random matrix theory via the asymptotic analysis of orthogonal polynomials. (Got all that? Go ask Percy Deift if that doesn't make any sense to you... I audited his class on some of this stuff when he visited Caltech this year, and it's pretty serious shit.)