arXiv subjects

Mathieu Merle

Publications and source records attributed to Mathieu Merle.

7 recordsLinked to original sources

arXiv2018

Cutoff for the mean-field zero-range process

We study the mixing time of the unit-rate zero-range process on the complete graph, in the regime where the number $n$ of sites tends to infinity while the density of particles per site stabilizes to some limit $\rho>0$. We prove that the worst-case total-variation distance to equilibrium drops abruptly from $1$ to $0$ at time $n\left(\rho+\frac{1}{2}\rho^2\right)$. More generally, we determine the mixing time from an arbitrary initial configuration. The answer turns out to depend on the largest initial heights in a remarkably explicit way. The intuitive picture is that the system separates into a slowly evolving solid phase and a quickly relaxing liquid phase. As time passes, the solid phase {dissolves} into the liquid phase, and the mixing time is essentially the time at which the system becomes completely liquid. Our proof combines meta-stability, separation of timescale, fluid limits, propagation of chaos, entropy, and a spectral estimate by Morris (2006).

math.PR↗

arXiv2015

Self-organized criticality in a discrete model for Smoluchowski's equation with limited aggregations

We introduce and study a discrete random model for Smoluchowski's equation with limited aggregations. The latter is a model of coagulation introduced by Bertoin which may exhibit gelation. In our model, a large number of particles are initially given a prescribed number of arms. These arms are activated independently after exponential times and successively linked together. However, when the size of a cluster goes above a fixed threshold, it falls instantaneously into the gel, meaning that it no longer interacts with the other particles in solution. The concentrations in this model asymptotically obey Smoluchowski's equation with limited aggregations. In this article, we study the discrete features of this model. We are able to argue that it remains closely related with a configuration model. We specifically obtain explicit expressions for the parameters of this configuration model, which show that it is subcritical before gelation, but remains critical afterwards. As a consequence, the asymptotic distribution of a typical cluster in solution is that of a subcritical Galton-Watson tree before gelation, while it is that of a critical Galton-Watson tree after gelation. Our model therefore exhibits self-organized criticality. Our study relies heavily on the study of the configuration model in the critical window. One of the main results of the paper is an extension of a result of Janson and Luczak on the size of the largest components in this critical window. Finally, as a consequence of our explicit expressions, we provide an explanation of analytic formulas established by Normand and Zambotti regarding the limiting concentrations in Smoluchowski's equation.

math.PR↗

arXiv2014

Self-organized criticality in a discrete model for Smoluchowski's equation

We study a discrete model of coagulation, involving a large number $N$ of particles. Pairs of particles are given i.i.d exponential clocks with parameter $1/N$. When a clock rings, a link between the corresponding pair of particles is created only if its two ends belong to small clusters, i.e. of size less than $\alpha(N)$, with $1 \ll \alpha(N) \ll N$. The concentrations of clusters of size $m$ in this model are known to converge as $N \to \infty$ to the solution to Smoluchowski's equation with a multiplicative kernel. Under the additional assumption $N^{2/3} \log^{\gamma}(N) \le \alpha(N)$, for some $\gamma > 1/3$, we study finer asymptotic properties of this model, namely the combinatorial structure of the graph consisting of small clusters. We prove that this graph is essentially an Erdos-Renyi random graph, which is subcritical before time 1, and remains critical after time 1. In particular, we show that our model exhibits self-organized criticality at a microscopic level: the limiting distribution of a typical finite cluster is that of a critical Galton-Watson tree. Our approach allows in particular to verify, under our additional assumption, a conjecture of Aldous.

math.PR↗

arXiv2013

Weak noise and non-hyperbolic unstable fixed points: Sharp estimates on transit and exit times

We consider certain one dimensional ordinary stochastic differential equations driven by additive Brownian motion of variance $\varepsilon ^2$. When $\varepsilon =0$ such equations have an unstable non-hyperbolic fixed point and the drift near such a point has a power law behavior. For $\varepsilon >0$ small, the fixed point property disappears, but it is replaced by a random escape or transit time which diverges as $\varepsilon \searrow0$. We show that this random time, under suitable (easily guessed) rescaling, converges to a limit random variable that essentially depends only on the power exponent associated to the fixed point. Such random variables, or laws, have therefore a universal character and they arise of course in a variety of contexts. We then obtain quantitative sharp estimates, notably tail properties, on these universal laws.

math.PR↗

arXiv2009

Scaling limit of the invasion percolation cluster on a regular tree

We prove existence of the scaling limit of the invasion percolation cluster (IPC) on a regular tree. The limit is a random real tree with a single end. The contour and height functions of the limit are described as certain diffusive stochastic processes. This convergence allows us to recover and make precise certain asymptotic results for the IPC. In particular, we relate the limit of the rescaled level sets of the IPC to the local time of the scaled height function.

math.PR↗

arXiv2007

A random walk on Z with drift driven by its occupation time at zero

We consider a nearest neighbor random walk on the one-dimensional integer lattice with drift towards the origin determined by an asymptotically vanishing function of the number of visits to zero. We show the existence of distinct regimes according to the rate of decay of the drift. In particular, when the rate is sufficiently slow, the position of the random walk, properly normalized, converges to a symmetric exponential law. In this regime, in contrast to the classical case, the range of the walk scales differently from its position.

math.PR↗

arXiv2006

Probability of hitting a distant point for the voter model started with a single one

The goal of this work is to find the asymptotics of the hitting probability of a distant point for the voter model on the integer lattice started from a single 1 at the origin. In dimensions 2 or 3, we obtain the precise asymptotic behavior of this probability. We use the scaling limit of the voter model started from a single 1 at the origin in terms of super-Brownian motion under its excursion measure. This invariance principle was stated by Bramson, Cox and Le Gall, as a consequence of a theorem of Cox, Durrett and Perkins. Less precise estimates are derived in dimensions greater than 4.

math.PR↗