Independent GUE minor processes of perfect matchings on rail-yard graphs

Zhongyang LiarXiv 2024 · DOI · arXiv

Abstract

This paper investigates perfect matchings on rail-yard graphs where the right boundary condition is an empty partition and the left boundary consists of alternating segments of removed or remaining vertices. By performing a quantitative analysis of Schur functions at general points, the study demonstrates that dimer locations near the right boundary converge in distribution to the spectra of independent Gaussian Unitary Ensemble (GUE) minor processes. This result extends the universality of GUE behavior to a broad class of bipartite graph geometries.

Introduction

In statistical mechanics, "dimer models"—perfect matchings of adjacent vertices in a graph—are used to understand physical systems ranging from molecular structures to crystal growth. A recurring theme in this field is the discovery that discrete combinatorial systems often behave like continuous random matrices in the large-scale limit. Specifically, near certain boundaries, the distribution of dimers often matches the spectra of the Gaussian Unitary Ensemble (GUE).

While this connection is well-established for simple grids like the hexagonal lattice or the Aztec diamond, it remained an open question for more generalized structures. This paper tackles rail-yard graphs, a highly general class of bipartite graphs, proving that even with complex piecewise boundary conditions, the system de-correlates into independent GUE processes.

A rail-yard graph with a perfect matching (dimer configuration). Thick lines represent the dimers.

Methodology

The research employs a combination of algebraic combinatorics and asymptotic probability. The core methodology involves mapping dimer configurations onto mathematical partitions, which are then analyzed using Schur processes.

Schur Functions and Particle-Hole Mapping

The author translates dimer locations into a "particle-hole" system. This representation allows the use of Schur functions—a powerful tool in symmetric function theory—to describe the probability distribution of the configurations. By introducing novel difference and differential operators, the author extracts the marginal distributions of dimers in specific regions of the graph.

Particle-hole configuration representing random partitions. Filled circles denote particles, and empty circles denote holes.

Asymptotic Factorization

A key technical hurdle was showing that the complex generating functions for these models factorize into independent components as the graph size grows. This was achieved through a quantitative analysis of a formula for Schur functions at general points (originally discovered by the author in 2018). This factorization allows multiple independent GUE processes to emerge from a single, interconnected graph.

Key Results

The primary contribution is Theorem 1.1, which establishes that under specific edge-weight conditions, the locations of dimers near the right boundary of a rail-yard graph converge in distribution to the spectra of independent GUE minor processes.

  • Weight-Driven Tangency: Unlike models where the geometry dictates the "tangent points" (where GUE behavior appears), this paper shows that varying edge weights alone can create these points, allowing for multiple independent processes to emerge on a single graph.
  • Universality: Despite the intricate "rail-yard" structure, the local behavior near the boundary remains universal, naturally de-correlating into independent components as the graph size goes to infinity.
  • Generalization to $n$ Processes: The framework applies to an arbitrary finite number ($n$) of independent GUE processes, extending previous results that typically handled only one or two.

Conclusion and Future Work

This work reinforces the idea that Schur functions are the natural language for connecting discrete tilings to the continuous world of random matrix theory. However, the current proof requires the right boundary to be an "empty partition" and relies on specific edge-weight assumptions (Assumption 2.19). Future work could explore more complex boundary conditions or more homogeneous weight distributions. Additionally, investigating which rail-yard configurations might lead to other ensembles, such as the Gaussian Orthogonal Ensemble (GOE), remains an open path.

Summary updated

Abstract

We study perfect matchings on the rail-yard graphs in which the right boundary condition is given by the empty partition and the left boundary can be divided into finitely many alternating line segments where all the vertices along each line segment are either removed or remained. When the edge weights satisfy certain conditions, we show that the distributions of the locations of certain types of dimers near the right boundary converge to the spectra of independent GUE minor processes. The proof is based on new quantitative analysis of a formula to compute Schur functions at general points discovered in \cite{ZL18}.
BibTeX
@misc{li2024independent,
  author = {Zhongyang Li},
  title = {{Independent GUE minor processes of perfect matchings on rail-yard graphs}},
  year = {2024},
  eprint = {2407.14626},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2407.14626},
}
Figure