Speaker: Doron Puder, TAU
Title: Aldous' spectral gap conjecture for normal sets
Aldous' spectral gap conjecture, proved in 2009 by Caputo, Liggett and Richthammer, states the following a priori very surprising fact: the spectral gap of a random walk on a finite graph is equal to the spectral gap of the interchange process on the same graph.
Speaker: Shira Zerbib (U. Michigan, Iowa State University)
Title: Envy-free division of a cake without the “hungry players" assumption
The fair division theorem due to Stromquist (1980) ensures that under some conditions it is possible to divide a rectangular cake into n pieces and assign one piece to each of n players such that no player strictly prefers a piece that has not been assigned to him.
Abstract: Any birational geometer would agree that the best algorithm
for resolution of singularities should run by defining a simple invariant of
the singularity and iteratively blowing up its maximality locus.
The only problem is that already the famous example of Whitney umbrella
shows that this is impossible, and all methods following Hironaka had
to use some history and resulted in more complicated algorithms.
Nevertheless, in a recent work with Abramovich and Wlodarczyk we did
construct such an algorithm, and an independent description of a similar
Abstract: I will revisit old results on liftings mod p2 and decompositions of de Rham complexes in positive characteristic (Deligne-I.) at the light of relations recently discovered independently by Bhargav Bhatt and myself between cotangent complexes, de Rham-Witt, and derived de Rham complexes.
Abstract: A Markov chain over a finite state space is said to exhibit the total variation cutoff phenomenon if, starting from some Dirac measure, the total variation distance to the stationary distribution drops abruptly from near maximal to near zero. It is conjectured that simple random walks on the family of $k$-regular, transitive graphs with a two sided $\epsilon$ spectral gap exhibit total variation cutoff (for any fixed $k$ and $\epsilon). This is known to be true only in a small number of cases.
For a finitely generated subgroup H of the free group F_r, the Stallings graph of H is a finite combinatorial graph, whose edges are labeled by r letters (and their inverses), so that paths in the graphs correspond precisely to the words in H. Furthermore, there is a map between the graphs of two subgroups H and J, precisely when one is a subgroups of the other. Stallings theory studies the algebraic information which is encoded in the combinatorics of these graphs and maps.
Abstract: In this talk, I shall present a generalization of the lattice point counting problem for Euclidean balls in the context of a certain type of homogeneous groups, the so-called Heisenberg groups.
Geometric class field theory is an analog of the classical class field theory over function fields in which functions are replaced by sheaves. In the first part of my talk, I will formulate the result and explain its proof over C (the field of complex numbers).
In the second part of the talk, I will try to outline the proof in the case of finite fields and indicate how this result implies the classical unramified global class field theory over function fields.
Most of the talk will be independent of the first one.
The talk will be based on work done by Furstenberg, taken mainly from his paper "Randon Walks and Discrete Subgroups of Lie Groups". We will present the idea of a boundary attached to a random walk on a group, and explain intuitively how it can be applied to prove that SL2(R) and SLn(R) - for n greater than 2 - do not have isomorphic lattices. Then we focus on a key step in that proof: Constructing a random walk on a lattice in SLn(R) that has the same boundary as a "spherical" random walk on SLn(R) itself.
Abstract: After reviewing the definition and the basic properties of the de Rham-Witt complex for smooth schemes over a perfect field, I will discuss the new approach to the subject developed by Bhatt, Lurie and Mathew.
I will explain the main results and sketch work in progress on the problems raised by this theory.