U(N) Weingarten function, known in computing the U(N) link integral, is an essential ingredient in physics. Although fewer people pay attention to SU(N), the SU(N) Weingarten function is important in the lattice gauge theory and it differs from U(N) . In this talk, I will present the derivation of the SU(N) Weingarten function using character theory and emphasize some details about how it differs from the perspective of polynomial representation of $GL_N$. We will also explore the nice combinatorial interpretation of the 1/N expansion of the Weingarten function using Hurwitz-Cayley graph which serves as the Feynman diagram

The notion of biexactness for groups was introduced by Ozawa in 2004 and has since become a major tool used for studying solidity of von Neumann algebras. In joint work with Changying Ding, we extended this notion from the group theory setting to the setting of von Neumann algebras, thereby giving a unified setting for proving solidity type results. We will discuss biexactness and solidity and give examples of solid von Neumann algebras that are not biexact.  

This talk presents a particle-based optimization method designed for addressing minimization problems with equality constraints, particularly in cases where the loss function exhibits non-differentiability or non-convexity. A rigorous mean-field limit of the particle system is derived, and the convergence of the mean-field limit to the constrained minimizer is established.

A 2015 conjecture of Ross and Yong proposes a K-Kohnert rule for Grothendieck polynomials. In this talk we discuss the utility of Kohnert rules then prove a special case of the Ross-Yong conjecture. We then show the conjecture fails in general.

This talk presents a particle-based optimization method designed for addressing global optimization problems, particularly in cases where the loss function exhibits non-differentiability or non-convexity. Numerically, we show that it outperforms gradient-based method in finding global optimizer. Theoretically, A rigorous mean-field limit of the particle system is derived, and the convergence of the mean-field limit to the global minimizer is established. In addition, we will talk about its application to the constrained optimization problems and federated learning.

We consider a synchronous process of particles moving on the vertices of a graph $G$, introduced by Cooper, McDowell, Radzik, Rivera and Shiraga (2018). Initially, $M$ particles are placed on a vertex of $G$. In subsequent time steps, all particles that are located on a vertex inhabited by at least two particles jump independently to a neighbour chosen uniformly at random. The process ends at the first step when no vertex is inhabited by more than one particle; we call this (random) time step the dispersion time.

In this work we study the case where $G$ is the complete graph on $n$ vertices and the number of particles is $M=n/2+\alpha n^{1/2} + o(n^{1/2})$, $\alpha\in \mathbb{R}$.This choice of $M$ corresponds to the critical window of the process, with respect to the dispersion time.

We show that the dispersion time, if rescaled by $n^{-1/2}$, converges in $p$-th mean, as $n\rightarrow \infty$ and for any $p \in \mathbb{R}$, to a continuous and almost surely positive random variable $T_\alpha$.

We find that $T_\alpha$ is the absorption time of a standard logistic branching process, thoroughly investigated by Lambert (2005), and we determine its expectation. In particular, in the middle of the critical window we show that $\mathbb{E}[T_0] = \pi^{3/2}/\sqrt{7}$, and furthermore we formulate explicit asymptotics when~$|\alpha|$ gets large that quantify the transition into and out of the critical window. We also study the random variable counting the \emph{total number of jumps} that are performed by the particles until the dispersion time is reached and prove that, if rescaled by $n\ln n$, it converges to $2/7$ in probability.

Based on joint work with Umberto De Ambroggio, Tamás Makai, and Annika Steibel; see arXiv:2403.05372

We will introduce an analytic notion of automorphic forms. These automorphic forms encode arithmetic data by way of their Fourier theory, and we will explore two different families of automorphic forms which have interesting congruences between their Fourier coefficients.

In mathematics in general, it is fruitful to allow randomness. Indeed, it is often easier to deal with random rather than deterministic objects. It seems miraculous, however, when we are able to say more about deterministic objects by treating them as random ones.

This idea applies in particular to the theory of discrete subgroups of Lie groups and locally symmetric manifolds.

The theory of invariant random subgroups (IRS), which has been developed quite rapidly during the last decade, has been very fruitful to the study of lattices and their asymptotic invariants. However, restricting to invariant measures limits the scope of problems that one can approach (in particular since the groups involved are highly non-amenable). It was recently realised that the notion of stationary random subgroups (SRS) is still very effective and opens paths to deal with questions which were thought to be unreachable.

In the talk I will describe various old and new results concerning arithmetic groups and general locally symmetric manifolds of finite as well as infinite volume that can be proved using `randomness', e.g.:

1. Kazhdan-Margulis minimal covolume theorem.

2. Most hyperbolic manifolds are non-arithmetic (a joint work with A. Levit).

3. Higher rank manifolds of large volume have a large injectivity radius (joint with Abert, Bergeron, Biringer, Nikolov, Raimbault and Samet).

4. Higher rank manifolds of infinite volume have infinite injectivity radius --- conjectured by Margulis (joint with M. Fraczyk).

What is the longest arithmetic progression that is a subset of a geometric progression? This problem is not as benign as it looks. But I bet we could do something or the other...