Professor: L. Williams (Science Center 510, e-mail williams@math.harvard.edu)
Office Hours: Thursdays 4pm or by appointment.
This seminar is intended to illustrate how research in mathematics actually progresses, using recent examples from the field of algebraic combinatorics. We will learn about the search for and discovery of a proof of a formula conjectured by Mills-Robbins-Rumsey in the early 1980's: the number of n by n alternating sign matrices. Alternating sign matrices are a curious family of mathematical objects, generalizing permutation matrices, which arise from an algorithm for evaluating determinants discovered by Charles Dodgson (better known as Lewis Carroll). They also have an interpretation as two-dimensional arrangements of water molecules, and are known in statistical physics as square ice. Although it was soon widely believed that the Mills-Robbins-Rumsey conjecture was true, the proof was elusive. Researchers working on this problem made connections to invariant theory, partitions, symmetric functions, and the six-vertex model of statistical mechanics. Finally in 1995 all these ingredients were brought together when Zeilberger and subsequently Kuperberg gave two different proofs of the conjecture. In this seminar we will survey these developments. If time permits, we will also get a glimpse of very recent activity in the field, for example the Razumov-Stroganov conjecture (now Cantini-Sportiello theorem).
The course will be fairly fast-paced; familiarity with proofs, and some basic notions from linear algebra will be helpful (for example the notion of the determinant of an n by n matrix).
A main reference for the course will be this book by David Bressoud, which you can also access here. We will supplement this book with various articles including Kuperberg's paper.
There is various software for playing with ASM's, including this code written by Dan Romik. (There's a Mathematica notebook and a Mac app.)Here are some resources for writing a math paper, and here are some resources for writing in Latex.