🔢 mathematics

Discrete $N$ -particle systems at high temperature through Jack generating functions

This paper establishes necessary and sufficient conditions for the Law of Large Numbers in discrete $N$ -particle systems at high temperature by utilizing Jack generating functions and the asymptotics of Cherednik operators, thereby unifying and extending previous results on quantized free cumulants and random matrix eigenvalues.

Original authors: Cesar Cuenca, Maciej Dołęga

Published 2026-10-20

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Cesar Cuenca, Maciej Dołęga

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a vast crowd of tiny particles, each pushing and pulling on its neighbors, trying to find a comfortable spot in a line. In the world of physics, these are not just abstract dots but real entities, like atoms in a gas or electrons in a metal, whose collective behavior determines the properties of the material they form. For decades, scientists have been fascinated by how these groups behave when they are very large. When the particles are cold and tightly packed, they interact strongly, creating complex, ordered patterns that resemble the spectra of random matrices. But what happens when the temperature rises? As the heat increases, the particles move faster, the interactions weaken, and the system begins to look more like a collection of independent individuals. This transition from a tightly knit community to a loose gathering is known as the high-temperature regime. Understanding exactly how the crowd settles down as it grows infinitely large and gets infinitely hot is a fundamental question in probability and statistical physics, one that bridges the gap between classical randomness and the more exotic rules of quantum-like independence.

A team of researchers has now solved a long-standing puzzle regarding this high-temperature behavior, but for a specific, discrete type of system that had previously resisted analysis. While the continuous versions of these particle systems were well understood, the discrete cases—where particles sit on a grid rather than floating freely in space—remained a mystery. The authors of this study have established precise conditions that predict exactly how these discrete groups will behave as they expand. They found that the key to unlocking the future of these systems lies in a specific mathematical tool called a Jack generating function. Think of this function not as a complex equation, but as a unique fingerprint or a master code that encodes the entire statistical history of the particle system. By examining the shape of this code at a specific point, the researchers can determine with certainty whether the system will settle into a predictable pattern as it grows.

The team proved that if this fingerprint meets two simple criteria, the system will obey a Law of Large Numbers. This means that as the number of particles becomes enormous, the average behavior of the group becomes stable and predictable, converging to a single, well-defined shape. If the fingerprint fails these criteria, the system remains chaotic and unpredictable. The researchers didn't just stop at finding the conditions; they also discovered a new way to calculate the exact shape of this final, stable state. They introduced a set of numbers, which they call quantized cumulants, that act like a new kind of statistical currency. Just as traditional currency can be exchanged for goods, these new numbers can be swapped, using a specific combinatorial recipe involving lattice paths, to reveal the final distribution of the particles. This recipe is a universal formula that works for a wide variety of systems, unifying different mathematical approaches that were previously thought to be separate.

To demonstrate the power of their method, the authors applied it to several real-world models, including systems of non-intersecting particles that move like random walkers on a line. These models are used to describe everything from traffic flow to the growth of crystals. In the high-temperature limit, the researchers showed that the behavior of these complex, interacting systems can be understood as a simple addition of their individual statistical properties. They proved that the final shape of the particle cloud is determined by adding up these new quantized numbers, a process that mirrors how independent random variables add up in classical probability, but with a twist that accounts for the specific discrete nature of the system. This work not only fills a gap in our understanding of discrete particle systems but also provides a robust toolkit for predicting the behavior of complex systems in the high-temperature limit, recovering known results for continuous systems as special cases while offering new insights for the discrete world.

Technical Summary: Discrete $N$ -Particle Systems at High Temperature Through Jack Generating Functions

Problem Statement
The paper addresses the global asymptotics of random discrete $N$ -particle systems on the real line as the number of particles $N$ tends to infinity. Specifically, it investigates the "high temperature" regime where the inverse temperature parameter $\theta$ (related to the standard $\beta$ -ensemble parameter by $2\theta = \beta$ ) tends to zero simultaneously with $N$ growing, such that the product $N\theta$ converges to a positive constant $\gamma$ .

While the limiting behavior of such systems is well-understood in the continuous case (e.g., $\beta$ -ensembles of random matrix eigenvalues) and in the fixed-temperature discrete case, the high-temperature regime for discrete systems had remained unexplored. The authors aim to establish necessary and sufficient conditions for the Law of Large Numbers (LLN) for the empirical measures of these discrete systems in this specific scaling limit.

Methodology
The authors employ the method of moments, utilizing a generating function approach tailored to discrete systems. The core technical tools include:

Jack Generating Functions: For a sequence of finite signed measures $\{P_N\}$ on $N$ -signatures (partitions or shifted integer tuples), the authors define the Jack generating function $G_{P_N, \theta}$ as a formal multivariate symmetric Laurent series involving Jack symmetric polynomials $P_\lambda$ .
Cherednik Operators: The analysis relies heavily on the action of Cherednik operators (differential-reflection operators) on these generating functions. The authors utilize the Hecke relations satisfied by these operators to derive asymptotic expansions.
Asymptotic Analysis: By studying the action of Cherednik operators on the logarithm of the Jack generating function in the limit $N \to \infty, N\theta \to \gamma$ , the authors extract the limiting moments.
Combinatorial Transforms: The relationship between the limiting moments and the coefficients of the generating function is expressed via a novel combinatorial transform involving Łukasiewicz paths (lattice paths with specific step types). This transform generalizes classical and free cumulant-moment relations.

Key Contributions and Results

Necessary and Sufficient Conditions for LLN: The main theorem (Theorem 3.6) establishes that a sequence of discrete $N$ -particle systems satisfies the LLN in the high-temperature regime if and only if the sequence is "HT-appropriate." This condition requires that the derivatives of the logarithm of the Jack generating function at the point $(1, \dots, 1)$ converge to specific limits (the quantized $\gamma$ -cumulants) while mixed derivatives vanish.
Quantized $\gamma$ -Cumulants: The paper introduces a new set of parameters, the quantized $\gamma$ -cumulants ( $\kappa^\gamma_n$ ), which uniquely determine the limiting measure. These cumulants are related to the limiting moments ( $m_\ell$ ) via a combinatorial formula (Theorem 1.2) expressed as a weighted sum over Łukasiewicz paths. This formula interpolates between classical cumulants ( $\gamma \to \infty$ ) and free cumulants ( $\gamma \to 0$ or specific normalizations).
Quantized $\gamma$ -Convolution: The authors define a convolution operation for probability measures based on the additivity of quantized $\gamma$ -cumulants. They prove that the Jack Littlewood–Richardson convolution of two sequences of measures corresponds to the addition of their quantized $\gamma$ -cumulants in the high-temperature limit. This provides a probabilistic interpretation for the structure constants of Jack polynomials in this regime.
Applications to Markov Chains: The framework is applied to a large class of Markov chains of $N$ non-intersecting particles with log-gas type interactions. The authors derive the LLN for these chains, showing that the limiting measure is the quantized $\gamma$ -convolution of the initial distribution and a specific "pure" measure determined by the specialization parameters.
Recovery of Known Results: The methods are robust enough to recover existing results:
- The fixed-temperature LLN results of Huang [Hua21] are recovered by taking the limit where $\theta$ is fixed.
- The quantized free cumulants of Bufetov and Gorin [BG15] (for $\theta=1$ ) are recovered.
- The high-temperature limits of continuous $\beta$ -ensembles (Benaych-Georges et al. [BGCG22]) are recovered as a degeneration of the discrete case.

Significance and Claims
The paper claims to fill a significant gap in the literature by providing the first comprehensive study of the high-temperature regime for discrete random particle systems. Unlike previous works that relied on analytic descriptions or specific continuous approximations, this work offers a combinatorial moment-cumulant formula that is universal across the discrete models considered.

The authors emphasize that their approach via Jack generating functions and Cherednik operators is distinct from the Dunkl operator/Bessel function methods used in continuous settings. They note that the lack of symmetry in Cherednik operators (compared to Dunkl operators) required new insights, specifically the use of Hecke relations to reduce generic terms to special ones.

The work connects representation theory (via Jack polynomials and Littlewood–Richardson coefficients) with probability theory (LLN and free probability deformations). It suggests that the "quantized $\gamma$ -convolution" and the associated semigroup framework (analogous to the free Brownian motion semigroup) provide a natural discrete counterpart to classical and free random matrix limits. The paper does not claim to solve the general positivity conjecture for Jack Littlewood–Richardson coefficients (Stanley's conjecture) but demonstrates that the quantized $\gamma$ -convolution exists for a large class of measures regardless of the full resolution of that conjecture.

More like this