arXiv Science⌕ Search

arXiv · 0707.4202

Multiclass Hammersley-Aldous-Diaconis process and multiclass-customer queues

Abstract

In the Hammersley-Aldous-Diaconis process infinitely many particles sit in R and at most one particle is allowed at each position. A particle at x$ whose nearest neighbor to the right is at y, jumps at rate y-x to a position uniformly distributed in the interval (x,y). The basic coupling between trajectories with different initial configuration induces a process with different classes of particles. We show that the invariant measures for the two-class process can be obtained as follows. First, a stationary M/M/1 queue is constructed as a function of two homogeneous Poisson processes, the arrivals with rate λand the (attempted) services with rate ρ>λ. Then put the first class particles at the instants of departures (effective services) and second class particles at the instants of unused services. The procedure is generalized for the n-class case by using n-1 queues in tandem with n-1 priority-types of customers. A multi-line process is introduced; it consists of a coupling (different from Liggett's basic coupling), having as invariant measure the product of Poisson processes. The definition of the multi-line process involves the dual points of the space-time Poisson process used in the graphical construction of the system. The coupled process is a transformation of the multi-line process and its invariant measure the transformation described above of the product measure.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Pablo A. Ferrari, James B. Martin. 2007-07-28. Multiclass Hammersley-Aldous-Diaconis process and multiclass-customer queues. https://arxiv.org/abs/0707.4202

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Dark Subspaces and Invariant Measures of Quantum Trajectories

Quantum trajectories are Markov processes describing the evolution of a quantum system subject to indirect measurements. They can be viewed as place dependent iterated function systems or the result of products of dependent and non identically distributed random matrices. In this article, we establish a complete classification of their invariant measures. The classification is done in two steps. First, we prove a Markov process on some linear subspaces called dark subspaces, defined in (Maassen, Kümmerer 2006), admits a unique invariant measure. Second, we study the process inside the dark subspaces. Using a notion of minimal family of isometries from a reference space to dark subspaces, we prove a set of measures indexed by orbits of a unitary group is the set of ergodic measures of quantum trajectories.

math.PR↗

Expected Signature Kernels for Rough Paths

The expected signature kernel is a similarity measure of probability laws on path space and enters directly into signature maximum mean discrepancies. Building on [P. K. Friz, P. P. Hager, and N. Tapia, Forum Math. Sigma, 10 (2022), e42] and [M. Lemercier, T. Lyons, and C. Salvi, SIAM J. Numer. Anal., 64 (2026), pp. 1095-1125], we characterize the expected signature kernel of group-valued inhomogeneous Lévy processes, aka Lévy rough paths, as the scalar component of the solution to an extended Goursat-type system. We give an original description, first in the continuous case where the finite-dimensional system is exact. In the presence of jumps, this system is inherently infinite-dimensional; we prove its well-posedness and derive finite-dimensional truncations with explicit error control. This permits direct evaluation of the reference-dependent terms in one-sample signature-MMD tests without an auxiliary reference sample. Finally, for conditionally Gaussian martingales, conditioning the usual unbiased empirical squared signature-MMD on the sampled quadratic variations yields an unbiased statistic for the same quantity with no larger variance. We derive computable finite-grid versions and illustrate the resulting variance reduction and gain in power.

math.PR↗

A central limit theorem for the range of random walk indexed by an infinite tree

We are interested in the second-order fluctuations of the size of the range of a random walk indexed by an infinite tree in ${\mathbb Z}^d$. Consider the Kesten tree with geometric offspring, we view the vertices on the left of its spine (the unique infinite path) as the past, and those on the right as the future. We show that the number of new sites in the future given its past has linear variance when $d>8$, and satisfies a central limit theorem (CLT) with Gaussian limiting distribution when $d>16$. The proof relies on the stationarity of the model under depth-first exploration, a general CLT by Dedecker and Merlevède [7], a truncation technique exploiting the local independence of tree structures, and a recursion argument for moment bounds.

math.PR↗