The two-periodic Aztec diamond is a dimer or random tiling model with three phases, solid, liquid and gas. The dimers form a determinantal point process with a somewhat complicated but explicit correlation kernel. I will discuss in some detail how the Airy point process can be found at the liquid-gas boundary by looking at suitable averages of height function differences. The argument is a rather complicated analysis using the cumulant approach and subtle cancellations. Joint work with Vincent Beffara and Sunil Chhita.[-]

A determinantal point process governed by a Hermitian contraction kernel $K$ on a measure space $E$ remains determinantal when conditioned on its configuration on a subset $B \subset E$. Moreover, the conditional kernel can be chosen canonically in a way that is "local" in a non-commutative sense, i.e. invariant under "restriction" to closed subspaces $L^2(B) \subset P \subset L^2(E)$. Using the properties of the canonical conditional kernel we establish a conjecture of Lyons and Peres: if $K$ is a projection then almost surely all functions in its image can be recovered by sampling at the points of the process.
Joint work with Alexander Bufetov and Yanqi Qiu.[-]

In this talk I will review the recent developments on weighted distances in scale free random graphs as well as highlight key techniques used in the proofs. We consider graph models where the degree distribution follows a power-law such that the empirical variance of the degrees is infinite, such as the configuration model, geometric inhomogeneous random graphs, or scale free percolation. Once the graph is created according to the model definition, we assign i.i.d. positive edge weights to existing edges, and we are interested in the proper scaling and asymptotic distribution of weighted distances.
In the infinite variance degree regime, a dichotomy can be observed in all these graph models: the edge weight distributions form two classes, explosive vs conservative weight distributions. When a distribution falls into the explosive class, typical distances converge in distribution to proper random variables. While, when a distribution falls into the conservative class, distances tend to infinity with the model size, according to a formula that captures the doubly-logarithmic graph distances as well as the precise behaviour of the distribution of edge-weights around the origin. An integrability condition decides into which class a given distribution falls.
This is joint work with Adriaans, Baroni, van der Hofstad, and Lodewijks.[-]

Markov chain Monte Carlo methods have become ubiquitous across science and engineering to model dynamics and explore large combinatorial sets. Over the last 20 years there have been tremendous advances in the design and analysis of efficient sampling algorithms for this purpose. One of the striking discoveries has been the realization that many natural Markov chains undergo phase transitions, whereby they abruptly change from being efficient to inefficient as some parameter of the system is modified. Generating functions can offer an alternative approach to sampling and they play a role in showing when certain Markov chains are efficient or not. We will explore the interplay between Markov chains, generating functions, and phase transitions for a variety of combinatorial problems, including graded posets, Boltzmann sampling, and 3-colorings on $Z^{2}$.[-]

Planar maps are planar graphs embedded in the sphere viewed modulo continuous deformations. There are two families of bijections between planar maps and lattice paths that are applied to prove scaling limit results of planar maps to so-called Liouville quantum gravity surfaces: metric bijections and mating-of-trees bijections. We will present scaling limit results obtained in this way, including works with Bernardi and Sun and with Albenque and Sun.[-]

In this talk, I will present recent results, obtained in collaboration with Laurent Ménard, about the geometry of spin clusters in Ising-decorated triangulations, and build on previously work obtained in collaboration with Laurent Ménard and Gilles Schaeffer.
In this model, triangulations are sampled together with a spin configuration on their vertices, with a probability biased by their number of monochromatic edges, via a parameter nu. The fact that there exists a combinatorial critical value for this model has been initially established in the physics literature by Kazakov and was rederived by combinatorial methods by Bousquet-Mélou and Schaeffer, and Bouttier, Di Francesco and Guitter.
Here, we give geometric evidence of that this model undergoes a phase transition by studying the volume and perimeter of its monochromatic clusters. In particular, we establish that, when nu is critical or subcritical, the cluster of the root is finite almost surely, and is infinite with positive probability for nu supercritical.[-]

The main purpose of this work is to provide a framework for proving that, given a family of random maps known to converge in the Gromov--Hausdorff sense, then some (suitable) conditional families of random maps converge to the same limit. As a proof of concept, we show that quadrangulations with a simple boundary converge to the Brownian disk. More precisely, we fix a sequence $(p_n)$ of even positive integers with $p_n\sim2\alpha \sqrt{2n}$ for some $\alpha\in(0,\infty)$. Then, for the Gromov--Hausdorff topology, a quadrangulation with a simple boundary uniformly sampled among those with $n$ inner faces and boundary length $p_n$ weakly converges, in the usual scaling $n^{-1/4}$, toward the Brownian disk of perimeter $3\alpha$.[-]

Bijections between planar maps and tree-like structures have been proven to play a crucial role in understanding the geometry of large random planar maps. Perhaps the most popular (and useful) bijections fit into two categories: bijections between maps and labeled trees and bijections between maps and blossoming trees. They were popularized in the late nineties in the important contribution of Schaeffer and they have been widely developed since then. It is natural to ask whether these bijections still hold when the underlying surface is no longer the sphere but any two-dimensional compact manifold? In this case trees are replaced by maps on a given surface with only one face and while the construction of Schaefer of the labeled-type bijection works independently on genus (but crucially depending on the assumption of orientability) his construction of the blossoming-type bijection was known only in the planar case. We will discuss a (recent?) development of these bijections that extends them to all compact two-dimensional manifolds. I will quickly review my previous joint work with Chapuy and its extension due to Bettinelli which treats the labeled-type bijection and will focus on a more recent work joint with Lepoutre which extends the blossoming-type bijection to non-oriented surfaces.[-]

In this talk, based on joint work with Alexandre Stauffer, I will consider the problem of providing 'uniform growth schemes' for various types of planar maps. In particular, we will discuss how to couple a uniform map with n faces with a uniform map with n+1 faces in such a way that the smaller map is always obtained from the larger by collapsing a single face. We show that uniform growth schemes exist for rooted 2p-angulations of the sphere and for rooted simple triangulations.[-]