Law of Large Numbers and Central Limit Theorem for random sets of solitons of the focusing nonlinear Schrödinger equation

This paper establishes a Law of Large Numbers and a Central Limit Theorem for random configurations of $N$ solitons in the focusing nonlinear Schrödinger equation, demonstrating that as $N$ increases, the random solution converges to a deterministic soliton gas limit with quantifiable fluctuations and correlation functions.

The Gist

The Big Picture: A Crowd of Solitary Waves

Imagine a calm ocean. Usually, if you throw a stone in, you get ripples that spread out and fade away. But in a special kind of water (described by the Focusing Nonlinear Schrödinger Equation, or fNLS), waves can behave differently. They can form "solitons"—these are like perfect, self-contained energy packets that travel forever without losing their shape or fading out. Think of them as indestructible, solitary surfers riding a wave that never breaks.

Usually, scientists study these solitons one by one, or in small, predictable groups. But in this paper, the authors ask: What happens if you have a massive, chaotic crowd of these solitons, all created by random chance?

The Setup: The "Soliton Gas"

The authors imagine a scenario where they generate N (a very large number) of these solitons.

• The Randomness: They don't pick the solitons' positions or speeds carefully. Instead, they use a "dice roll" (random probability) to decide where each soliton's "eigenvalue" (a number that determines its speed and shape) comes from.
• The Gas: As N gets bigger and bigger, these individual solitons start to look less like distinct surfers and more like a dense gas or a fog of waves.

The paper asks two main questions about this "Soliton Gas":

1. The Law of Large Numbers: If we have a huge crowd, does the chaotic mess settle down into a predictable, smooth pattern?
2. The Central Limit Theorem: If there are tiny, random wiggles left over after the pattern settles, do those wiggles follow a familiar bell-curve distribution (like heights in a population)?

The Analogy: The "Average" Wave vs. The "Real" Wave

To understand the math, imagine a classroom full of students (the solitons).

• The Real Situation ($\psi_N$): Every student is shouting a different note at a slightly different volume. The total sound in the room is a chaotic, fluctuating roar. This is the random N-soliton solution.
• The Average Situation ($\psi_\infty$): Imagine you take a microphone, record the room, and calculate the "average" sound wave. This creates a smooth, predictable hum. This is the deterministic solution the authors construct.

The authors prove that as the number of students (solitons) goes to infinity:

1. The Roar Becomes a Hum: The chaotic sound of the real room gets closer and closer to the smooth average hum. The difference between the two becomes negligible. This is the Law of Large Numbers.
2. The Wiggles are Normal: If you look at the tiny differences between the real roar and the average hum, those differences aren't random chaos; they follow a very specific, predictable statistical pattern (a Gaussian distribution). This is the Central Limit Theorem.

How They Did It: The "Error" Detective

The math behind this is tricky because the waves interact with each other in complex, non-linear ways (they crash into each other and change shape). You can't just add them up like simple numbers.

The authors used a powerful mathematical tool called the Inverse Scattering Transform. Think of this as a magic decoder ring.

• The Problem: Solving the wave equation directly is like trying to untangle a knot of 1,000 ropes while they are moving.
• The Trick: The decoder ring translates the moving, tangled ropes into a set of simple, static numbers (the "scattering data"). In this "number world," the waves don't interact; they just evolve linearly (like a clock ticking).
• The Randomness: The authors put their randomness into these static numbers.
• The Comparison: They compared the "number world" of the chaotic crowd against the "number world" of the smooth average. They proved that the "error" (the difference between the two) shrinks to zero as the crowd gets larger.

The Key Findings

1. Predictability from Chaos: Even though the starting conditions were completely random, the resulting "Soliton Gas" behaves in a highly predictable, smooth way when you look at it on a large scale.
2. The "Soliton Gas" is Real: They confirmed that the theoretical concept of a "soliton gas" (a dense collection of interacting solitons) actually exists mathematically and can be described by a specific smooth solution ($\psi_\infty$).
3. Fluctuations are Under Control: They didn't just say the average is right; they calculated exactly how much the random version wobbles around that average. They found these wobbles follow a standard bell curve, meaning we can predict the probability of extreme deviations.

What This Means (Without Speculation)

The paper provides a rigorous mathematical proof that randomness in the starting ingredients leads to order in the final result for these specific types of waves. It bridges the gap between the microscopic world of individual, colliding solitons and the macroscopic world of smooth, predictable wave patterns.

In short: If you throw enough random solitons into a pot, they will eventually cook into a perfectly smooth soup, and we can now mathematically prove exactly how smooth that soup will be and how much it might wiggle.

Technical Summary

Technical Summary: Law of Large Numbers and Central Limit Theorem for Random Sets of Solitons of the Focusing Nonlinear Schrödinger Equation

Problem Statement
The paper addresses the statistical behavior of solutions to the cubic focusing Nonlinear Schrödinger (fNLS) equation in one space dimension:
$$i\psi_t + \frac{1}{2}\psi_{xx} + |\psi|^2\psi = 0, \quad x \in \mathbb{R}, t \in \mathbb{R}^+.$$
While the evolution of random initial data for linear equations is well-understood via superposition of uncorrelated waves, the nonlinear case presents significant challenges. For nonlinear waves, the probability distribution of the wave field deforms substantially over time. Previous work has largely focused on weakly nonlinear regimes described by wave kinetic equations. However, for large-amplitude, fundamentally nonlinear solutions such as solitons, a rigorous statistical theory is lacking.

The authors investigate $N$-soliton solutions where the spectral data (eigenvalues and norming constants) are randomized. Specifically, the eigenvalues $\{\lambda_k\}_{k=1}^N$ are independent and identically distributed (i.i.d.) random variables sampled from a probability distribution with compact support in the upper half-plane $\mathbb{C}_+$. The norming constants are determined by a smooth interpolating function of these eigenvalues. The central question is whether, as $N \to \infty$, the random $N$-soliton solution converges to a deterministic limit (a "soliton gas") and whether the fluctuations around this limit follow a Central Limit Theorem (CLT).

Methodology
The analysis relies on the integrability of the fNLS equation via the Inverse Scattering Transform (IST). The solution is reconstructed from scattering data using a Riemann-Hilbert (RH) problem.

1. Random RH Formulation: The authors formulate the $N$-soliton solution as a random meromorphic RH problem. To facilitate asymptotic analysis, they convert the pole conditions (associated with discrete eigenvalues) into jump relations across smooth contours $\gamma_+$ and $\gamma_-$ enclosing the support of the eigenvalue distribution. This yields a random jump matrix $J_N(z; x, t, \lambda)$.
2. Averaged Deterministic Limit: A deterministic "averaged" RH problem is constructed by taking the expectation of the random jump matrix, $J(z; x, t) = \mathbb{E}[J_N(z; x, t, \lambda)]$. Due to the i.i.d. nature of the eigenvalues and the interpolation of norming constants, this averaged jump matrix is independent of $N$. The solution to this deterministic RH problem defines a limiting function $\psi_\infty(x, t)$, interpreted as a soliton gas solution.
3. Error Analysis: The difference between the random solution $\psi_N$ and the deterministic limit $\psi_\infty$ is analyzed by studying the error matrix $E(z) = M_N(z)M(z)^{-1}$, where $M_N$ and $M$ are the solutions to the random and averaged RH problems, respectively.
4. Small Norm Argument: The jump matrix for the error problem, $J_E$, is shown to be a perturbation of the identity matrix, $J_E = I + W_N$. The term $W_N$ depends on linear statistics of the random eigenvalues. Using probabilistic estimates (specifically, bounds on the linear statistics $X_N^f$), the authors demonstrate that with high probability, the norm of $W_N$ is small as $N \to \infty$. This allows the error matrix $E$ to be expanded in a convergent Neumann series.
5. Probabilistic Limits:
   • Law of Large Numbers (LLN): By bounding the error terms in the Neumann series and controlling the probability of "bad" configurations (where the norm is not small), the authors prove convergence in mean ($L^1$) of $\psi_N$ and $|\psi_N|^2$ to $\psi_\infty$ and $|\psi_\infty|^2$.
   • Central Limit Theorem (CLT): The leading order term in the expansion of the scaled difference $\sqrt{N}(\psi_N - \psi_\infty)$ is identified as a linear statistic of the random eigenvalues. By the classical CLT, this term converges to a Gaussian random variable. The higher-order terms in the Neumann series are shown to be negligible in probability.

Key Contributions and Results

• Existence of a Deterministic Soliton Gas: The paper establishes the existence of a unique classical solution $\psi_\infty(x, t)$ to the fNLS equation derived from the expectation of the spectral data. This solution is interpreted as the macroscopic limit of a random soliton gas.
• Law of Large Numbers (Theorem 2.6): The authors prove that for $(x, t)$ in a compact set of $\mathbb{R} \times \mathbb{R}^+$, the random $N$-soliton solution $\psi_N(x, t; \lambda)$ and its modulus squared converge in mean to the deterministic solution $\psi_\infty(x, t)$ and $|\psi_\infty(x, t)|^2$ respectively:
  $$\lim_{N \to \infty} \mathbb{E}[|\psi_N - \psi_\infty|] = 0, \quad \lim_{N \to \infty} \mathbb{E}[||\psi_N|^2 - |\psi_\infty|^2|] = 0.$$
• Central Limit Theorem (Theorem 2.7): The paper proves that the fluctuations around the mean, scaled by $\sqrt{N}$, converge in distribution to Gaussian random variables. Specifically, $\sqrt{N}(\psi_N - \psi_\infty)$ converges to a complex Gaussian variable $X_{G1}$, and $\sqrt{N}(|\psi_N|^2 - |\psi_\infty|^2)$ converges to a real Gaussian variable $X_{G2}$. Both have zero mean, and their variances are explicitly computed as integrals involving the solution of the averaged RH problem and the interpolating function.
• Correlation Functions (Theorem 2.8): The authors compute the two-point correlation functions for the fluctuations, showing they converge to the covariance of the limiting Gaussian variables.

Significance and Claims
The paper claims to provide the first rigorous derivation of kinetic theory results for random soliton gases in the context of the focusing NLS equation via the analysis of the underlying nonlinear PDE. While kinetic theories (Generalized Hydrodynamics) have been proposed and numerically verified for soliton gases, this work offers a rigorous mathematical foundation for the Law of Large Numbers and Central Limit Theorem in this setting.

The authors emphasize that their approach introduces randomness in the linear setting of the scattering data, which remains uncorrelated and unchanged in distribution as the solution evolves, allowing for the study of random large-amplitude nonlinear dynamics. The results bridge the gap between the microscopic description of individual solitons and the macroscopic statistical behavior of soliton gases, providing a predictive statistical theory for large-amplitude waves over large scales of space and time. The work is presented as a rigorous step toward understanding the statistical mechanics of integrable nonlinear dispersive systems.