Girko matrices have independent and identically distributed entries of mean zero and unit variance. In this note, we consider the random matrix model formed by the ratio of two independent Girko matrices, its entries are dependent and heavy-tailed. Our main message is that divided by the square root of the dimension, the spectral radius of the ratio converges in distribution, when the dimension tends to infinity, to a universal heavy-tailed distribution. We provide a mathematical proof of this high-dimensional phenomenon, under a fourth moment matching with a Gaussian case known as the complex Ginibre ensemble. In this Gaussian case, the model is known as the spherical ensemble, and its spectrum is a determinantal planar Coulomb gas. Its image by the inverse stereographic projection is a rotationally invariant gas on the two-sphere. A crucial observation is the invariance in law of the model under inversion, related to its spherical symmetry, and that makes, in a sense, edge and bulk equivalent. Our approach involves Girko Hermitization, local law estimates for Wigner matrices, lower bound estimates on the smallest singular value, and convergence of kernels of determinantal point processes. The universality of the high-dimensional fluctuation of the spectral radius of the ratio of Girko matrices turns out to be remarkably more accessible mathematically than for a single Girko matrix!

This paper studies the clustering behavior of weakly interacting diffusions under the influence of sufficiently localized attractive interaction potentials on the one-dimensional torus. We describe how this clustering behavior is closely related to the presence of discontinuous phase transitions in the mean-field PDE. For local attractive interactions, we employ a new variant of the strict Riesz rearrangement inequality to prove that all global minimizers of the free energy are either uniform or single-cluster states, in the sense that they are symmetrically decreasing. We analyze different timescales for the particle system and the mean-field (McKean-Vlasov) PDE, arguing that while the particle system can exhibit coarsening by both coalescence and diffusive mass exchange between clusters, the clusters in the mean-field PDE are unable to move and coarsening occurs via the mass exchange of clusters. By introducing a new model for this mass exchange, we argue that the PDE exhibits dynamical metastability. We conclude by presenting careful numerical experiments that demonstrate the validity of our model.

In finite dimension, the long-time and metastable behavior of a gradient flow perturbated by a small Brownian noise is well understood. A similar situation arises when a Wasserstein gradient flow over a space of probability measure is approximated by a system of mean-field interacting particles, but classical results do not apply in these infinite-dimensional settings. This work is concerned with the situation where the objective function of the optimization problem contains an entropic penalization, so that the particle system is a Langevin diffusion process. We consider a very simple class of models, for which the infinite-dimensional behavior is fully characterized by a finite-dimensional process. The goal is to have a flexible class of benchmarks to fix some objectives, conjectures and (counter-)examples for the general situation. Inspired by the systematic study of these toy models, one application is presented on the continuous Curie-Weiss model in a symmetric double-well potential. We show that, at the critical temperature, although the $N$-particle Gibbs measure does not satisfy a uniform-in-$N$ standard log-Sobolev inequality (the optimal constant growing like $\sqrt{N}$), it does satisfy a more general Lojasiewicz inequality uniformly in $N$, inducing uniform polynomial long-time convergence rates, propagation of chaos at stationarity and uniformly in time, and creation of chaos.

The two-star random graph is the simplest exponential random graph model with nontrivial interactions between the graph edges. We propose a set of auxiliary variables that control the thermodynamic limit where the number of vertices N tends to infinity. Such 'master variables' are usually highly desirable in treatments of 'large N' statistical field theory problems. For the dense regime when a finite fraction of all possible edges are filled, this construction recovers the mean field solution of Park and Newman, but with an explicit control over the 1/N corrections. We use this advantage to compute the first subleading correction to the Park-Newman result, which encodes the finite, nonextensive contribution to the free energy. For the sparse regime with a finite mean degree, we obtain a very compact derivation of the Annibale-Courtney solution, originally developed with the use of functional integrals, which is comfortably bypassed in our treatment.

We study the critical behavior of the dissipative abelian sandpile model on Z with dissipative sites at arbitrary positions (x_k). This is equivalent to studying whether the expected stopping time of a trapped random walk on Z is finite. Our main contribution is a precise description of this phase transition via three distinct regimes. Our analysis captures criticality through the asymptotic behavior of an explicit recursive sequence, revealing counterintuitive phenomena in which traps may be farther apart yet the stopping time becomes shorter.

For Markov kernels $P$ on a general state space $\mathcal{X}$, we introduce a new class of averaged Markov kernels $P_{da}(G,\nu)$ of $P$ induced by a group $G$ that acts on $\mathcal{X}$ and a probability measure $\nu$ on $G \times G$. Notable special cases are the group-orbit average $\overline{P}$, left-average $P_{la}$, right-average $P_{ra}$ and the independent-double-average $(P_{la})_{ra}$. For $\pi$-stationary $P$ in which $\pi$ is invariant with respect to $G$, we show that in general $P_{da}$ enjoys favorable convergence properties than $P$ based on metrics such as spectral gap or asymptotic variance, and within the family of $P_{da}$ the most preferable kernel is in general $(P_{la})_{ra}$. We demonstrate that $P_{la}, P_{ra}, (P_{la})_{ra}$ are comparable in terms of mixing times, which supports the use of $P_{la}, P_{ra}$ in practice as computationally cheaper alternatives over $(P_{la})_{ra}$. These averaged kernels also admit natural geometric interpretations: they emerge as unique projections of $P$ onto specific $G$-invariant structures under the Kullback-Leibler divergence or the Hilbert-Schmidt norm and satisfy Pythagorean identities. On the other hand, in the general case if $\pi$ is not invariant with respect to $G$, we propose and study a technique that we call state-dependent averaging of Markov kernels which generalizes the earlier results to this setting. As examples and applications, this averaging perspective not only allows us to recast state-of-the-art Markov chain samplers such as Hamiltonian Monte Carlo or piecewise-deterministic Markov processes as specific cases of $P_{da}$, but also enables improvements to existing samplers such as Metropolis-Hastings, achieving rapid mixing in some toy models or when $\pi$ is the discrete uniform distribution.

We study the trade-off between envy and inefficiency in repeated resource allocation settings with stochastic replenishments, motivated by real-world systems such as food banks and medical supply chains. Specifically, we consider a model in which a decision-maker faced with stochastic demand and resource donations must trade off between an equitable and efficient allocation of resources over an infinite horizon. The decision-maker has access to storage with fixed capacity $M$, and incurs efficiency losses when storage is empty (stockouts) or full (overflows). We provide a nearly tight (up to constant factors) characterization of achievable envy-inefficiency pairs. Namely, we introduce a class of Bang-Bang control policies whose inefficiency exhibits a sharp phase transition, dropping from $\Theta(1/M)$ when $\Delta = 0$ to $e^{-\Omega(\Delta M)}$ when $\Delta > 0$, where $\Delta$ is used to denote the target envy of the policy. We complement this with matching lower bounds, demonstrating that the trade-off is driven by supply, as opposed to demand uncertainty. Our results demonstrate that envy-inefficiency trade-offs not only persist in settings with dynamic replenishment, but are shaped by the decision-maker's available capacity, and are therefore qualitatively different compared to previously studied settings with fixed supply.

Consider a generalized Elephant Random Walk in which the step is chosen by selecting $k$ previous steps with $k$ odd and then going in the majority direction with a probability $p$ and in the opposite direction otherwise. In the $k=1$ case the model is the original one and could be resolved exactly by analogy with Friedman's urn. However the analogy cannot be extended to the $k>2$ case already. In this paper we show how to treat the model for each $k$ by analogy with the more general urn model of Hill, Lane and Sudderth. Interestingly for $k>2$ we found a critical dependence from the initial conditions beyond a certain values of the memory parameter $p$, and regions of convergence with entropy that is sub-linear in the number of steps.

Motivated by the dissipative abelian sandpile model, we analyze the trajectories of a one-dimensional random walk in a landscape of soft traps. These traps, placed at increasing distances from each other, correspond to dissipative sites in the associated dissipative abelian sandpile model. We identify a critical growth rate of the sizes of intervals between successive traps where there is a transition between finiteness and non-finiteness of the expected survival time of the random walk. This corresponds to a transition between non-criticality and criticality of the associated dissipative abelian sandpile model. Therefore, in this setting, we thus identify precisely how much dissipation can be added to the original abelian sandpile model in order to disrupt its criticality.

In this paper we show how to extend the Sample-Path Large Deviation Principle for the urn model of Hill, Lane and Sudderth to the case in which the increment of the urn is not a binary variable. In particular, we sketch how to modify the Theorem 1 given in [Stochastic Processes and their Applications 127 (2017) 3372-3411] to include also urn processes with increments taking more than two values.

This paper establishes a Transition Path Theory (TPT) for L\'{e}vy-type processes, addressing a critical gap in the study of the transition mechanism between meta-stabile states in non-Gaussian stochastic systems. A key contribution is the rigorous derivation of the stochastic differential equation (SDE) representation for transition path processes, which share the same distributional properties as transition trajectories, along with a proof of its well-posedness. This result provides a solid theoretical foundation for sampling transition trajectories. The paper also investigates the statistical properties of transition trajectories, including their probability distribution, probability current, and rate of occurrence.

Multi-agent reinforcement learning, despite its popularity and empirical success, faces significant scalability challenges in large-population dynamic games. Graphon mean field games (GMFGs) offer a principled framework for approximating such games while capturing heterogeneity among players. In this paper, we propose and analyze a policy optimization framework for continuous-time, finite-horizon linear-quadratic GMFGs. Exploiting the structural properties of GMFGs, we design an efficient policy parameterization in which each player's policy is represented as an affine function of their private state, with a shared slope function and player-specific intercepts. We develop a bilevel optimization algorithm that alternates between policy gradient updates for best-response computation under a fixed population distribution, and distribution updates using the resulting policies. We prove linear convergence of the policy gradient steps to best-response policies and establish global convergence of the overall algorithm to the Nash equilibrium. The analysis relies on novel landscape characterizations over infinite-dimensional policy spaces. Numerical experiments demonstrate the convergence and robustness of the proposed algorithm under varying graphon structures, noise levels, and action frequencies.

We consider multidimensional random walks in pyramidal cones (or multidimensional orthants), which are intersections of a finite number of half-spaces. We explore the connection between the existence of (positive) discrete harmonic polynomials for the random walks, with Dirichlet conditions on the boundary of the cone, and geometric properties of the cone, being or not the Weyl chamber of a finite Coxeter group. We prove that the first property implies the second, derive the converse in dimension two and show in this case that it coincides with the probabilistic harmonic function.

The optimal matching of point clouds in $\mathbb{R}^d$ is a combinatorial problem; applications in statistics motivate to consider random point clouds, like the Poisson point process. There is a crucial dependance on dimension $d$, with $d=2$ being the critical dimension. This is revealed by adopting an analytical perspective, connecting e.\,g.~to Optimal Transportation. These short notes provide an introduction to the subject. The material presented here is based on a series of lectures held at the International Max Planck Research School during the summer semester 2022. Recordings of the lectures are available at this https URL

Temporal difference (TD) learning is a foundational algorithm in reinforcement learning (RL). For nearly forty years, TD learning has served as a workhorse for applied RL as well as a building block for more complex and specialized algorithms. However, despite its widespread use, TD procedures are generally sensitive to step size specification. A poor choice of step size can dramatically increase variance and slow convergence in both on-policy and off-policy evaluation tasks. In practice, researchers use trial and error to identify stable step sizes, but these approaches tend to be ad hoc and inefficient. As an alternative, we propose implicit TD algorithms that reformulate TD updates into fixed point equations. Such updates are more stable and less sensitive to step size without sacrificing computational efficiency. Moreover, we derive asymptotic convergence guarantees and finite-time error bounds for our proposed implicit TD algorithms, which include implicit TD(0), TD($\lambda$), and TD with gradient correction (TDC). Our results show that implicit TD algorithms are applicable to a much broader range of step sizes, and thus provide a robust and versatile framework for policy evaluation and value approximation in modern RL tasks. We demonstrate these benefits empirically through extensive numerical examples spanning both on-policy and off-policy tasks.