Nekhoroshev type stability for non-local semilinear Schrödinger equations

This paper establishes the first rigorous Nekhoroshev-type stability results for ultra-differentiable solutions of non-local semilinear Schrödinger equations without external parameters by employing rational normal forms and a novel global vector field norm, achieving optimal stability times under Gevrey regularity assumptions.

The Gist

Imagine you are watching a complex dance floor filled with thousands of dancers. In the world of physics, these dancers are waves (specifically, quantum waves described by the Schrödinger equation).

Usually, when these waves interact, they can get chaotic. They might bump into each other, exchange energy, and eventually, the whole dance floor could turn into a messy, unpredictable riot. This is what mathematicians call "instability."

This paper is about proving that, under certain conditions, this dance floor can stay orderly and predictable for an incredibly long time—so long that it feels like "forever" to us humans.

Here is a breakdown of the paper's big ideas using simple analogies:

1. The Problem: The "Long-Distance" Dance

Most previous studies looked at dancers who only interact with their immediate neighbors (like a local party). But this paper looks at a non-local system.

• The Analogy: Imagine a dance where a dancer in the corner can "feel" and react to a dancer on the opposite side of the room, even if they are far apart. This is like a gravitational pull or a long-range magnetic force.
• The Challenge: When everyone can influence everyone else from across the room, the math gets incredibly messy. It's like trying to predict the weather when every single molecule in the atmosphere is talking to every other molecule instantly.

2. The Goal: The "Nekhoroshev" Promise

The authors are trying to prove Nekhoroshev stability.

• The Analogy: Think of a ball sitting in a deep, smooth valley. If you nudge it slightly, it rolls back to the center. In a chaotic system, a tiny nudge might eventually send the ball rolling out of the valley entirely.
• The Promise: The authors prove that for this specific type of dance, if you start with a small, gentle nudge (a small initial disturbance), the ball won't roll out of the valley for a time that is exponentially long.
  • If the nudge is size $0.001$, the system might stay stable for$10^{100}$ years. That is longer than the age of the universe.

3. The Secret Weapon: "Rational Normal Forms"

To prove this, the authors had to invent a new way of organizing the math. They call it the Rational Normal Form.

• The Old Way (The "Counting" Method): Imagine trying to organize a library by counting every single book on every single shelf, noting the exact color of the spine, and tracking how many books are in the top row vs. the bottom row. It's tedious, and if you miss one count, the whole system breaks. Previous methods were like this—they had to track the "degree" (complexity) of every interaction term meticulously.
• The New Way (The "Vector Field" Method): The authors introduced a new measuring stick (a "vector field norm").
  • The Analogy: Instead of counting every book, they just look at the "flow" of the library. They ask: "Is the general direction of the books moving toward chaos or order?"
  • Why it's better: This new tool ignores the messy details of counting individual books. It treats the whole interaction as a single, smooth flow. This allowed them to handle the "long-distance" interactions without getting lost in the math weeds.

4. The "Internal Parameter" Trick

Usually, to prove stability, mathematicians need to tweak the system slightly (like adding a tiny external weight to the dance floor) to make the math work. This is called an "external parameter."

• The Innovation: This paper proves stability without adding any external weights.
• The Analogy: Instead of bringing in a referee to fix the dance, the authors realized that the dancers themselves (the initial energy of the waves) act as their own referees. The size of the initial dance is the control knob. This is much harder to prove because the "referee" is part of the chaos, but the authors cracked the code.

5. The "Smoothness" of the Dancers

The paper looks at two types of dancers:

1. Gevrey Class: These dancers are very smooth, almost like silk.
2. Logarithmic Ultra-differentiable: These are even smoother, but in a very specific, mathematical way (think of them as having "super-silk" skin).

• The Result: The authors proved that even for these ultra-smooth dancers, the system remains stable for a massive amount of time. They matched a famous prediction made by mathematician Jean Bourgain years ago, confirming that the stability time is as long as theoretically possible.

6. The "Bad Seats" (Measure Estimate)

Is this stability true for every possible starting position? No.

• The Analogy: Imagine a stadium. Most seats offer a great view of the stable dance. However, there are a few "bad seats" (resonant zones) where, if you start there, the dance immediately turns chaotic.
• The Finding: The authors calculated exactly how many "bad seats" there are. They found that the bad seats are incredibly rare—like finding a single grain of sand in a beach. For 99.999...% of starting positions, the system will remain stable for eons.

Summary

In plain English, this paper says:

"We have a complex system where waves interact over long distances. Using a brand-new mathematical tool that simplifies how we track these interactions, we proved that if you start the system gently, it will stay calm and orderly for a time so long it's practically infinite. We did this without needing to cheat by adding external controls, and we showed that this holds true for almost every possible starting scenario."

This is a major step forward in understanding how complex quantum systems (like Bose-Einstein condensates or self-gravitating stars) behave over long periods without falling apart.

Technical Summary

Here is a detailed technical summary of the paper "Nekhoroshev type stability for non-local semilinear Schrödinger equations" by Bingqi Yu and Yong Li.

1. Problem Statement

The paper investigates the long-time stability of solutions to non-local semilinear Schrödinger equations on a torus $T^d$. The equation is given by:
$$i\partial_t u + \Delta u + u \int_{T^d} K(x-y)|u(y,t)|^2 dy = 0$$
where $K$ is a non-local interaction kernel. The authors focus on the Nekhoroshev stability problem in an infinite-dimensional setting without external parameters.

Key Challenges:

• Infinite Dimensions: The system has infinitely many degrees of freedom, leading to a dense set of resonances that complicates stability analysis.
• Internal Parameters: Unlike previous works that rely on external parameters (e.g., convolution potentials) to tune frequencies, this work treats the initial data amplitudes as internal parameters. This is significantly more difficult because these parameters are small and require refined non-resonance conditions.
• Regularity: The study extends beyond standard Sobolev or Gevrey classes to include logarithmic ultra-differentiable regularity, a class strictly between smooth and analytic functions.
• Non-locality: The interaction term involves a convolution kernel $K$, which can exhibit either polynomial decay ($K_k \sim |k|^{-p}$) or exponential decay ($K_k \sim e^{-|k|^\beta}$) in Fourier space.

2. Methodology

The authors employ a sophisticated combination of Rational Normal Forms (RNF) and measure-theoretic estimates.

• Rational Normal Forms: Instead of standard Birkhoff normal forms, the authors use the RNF framework (introduced by Bernier, Faou, and Grébert). This allows the homological equations to contain terms where the solution $u$ appears in the denominator (rational terms), effectively using the amplitudes of the initial data as internal parameters to resolve resonances.
• Novel Vector Field Norm: A critical methodological innovation is the introduction of a global vector field norm adapted to the rational normal form.
  • Traditional approach: Required tracking the degrees of polynomials in numerators and denominators separately, leading to complex combinatorial estimates.
  • New approach: The vector field norm applies directly to the vector field, unifying the treatment of polynomial and rational nonlinearities and eliminating the need for degree tracking during iterations.
• Truncation and Decomposition: The phase space is decomposed into low modes ($|k| \le M$) and high modes ($|k| > M$).
  • Low modes: Transformed into an integrable rational normal form.
  • High modes: Controlled via "truncation estimates" showing that high-mode interactions vanish to high order, preventing energy transfer to high frequencies.
• Bootstrap Argument: A stability time $T_r$ is established by assuming the solution grows beyond a certain bound and deriving a contradiction using the transformed Hamiltonian system.

3. Key Contributions

1. First Results without External Parameters: The paper establishes the first rigorous Nekhoroshev-type stability results for infinite-dimensional Hamiltonian systems where the non-resonance conditions rely solely on internal parameters (initial data amplitudes) rather than external tuning.
2. Ultra-Differentiable Regularity: The authors extend stability theory to logarithmic ultra-differentiable classes ($W^{U}_{s,\theta}$), a regularity class introduced by Z. Hani. This is the first time such regularity has been analyzed in the context of internal parameter stability.
3. Optimal Stability Times: Under Gevrey regularity assumptions, the authors achieve stability times matching Bourgain's conjectured optimal time (order $\exp(c |\ln r|^2 / \ln|\ln r|)$).
4. Improved Measure Estimates: The paper provides sharper estimates for the measure of the "bad" set (resonant set) of initial data. The measure of the unstable set is bounded by $r^{2/5}$, improving upon previous results (e.g., $r^{1/3}$ or $r^{1/35}$) for similar equations.
5. Unified Framework for Non-locality: The method handles both polynomial decay kernels (modeling long-range interactions like Schrödinger-Newton) and exponential decay kernels (modeling highly non-local media) within a single theoretical framework.

4. Main Results

The paper presents four main theorems covering combinations of two kernel types and two function spaces:

Function Spaces:

• Gevrey Class ($W^G_{s,g}$): Regularity $e^{s|j|^g}$ with $0 < g < 1$.
• Logarithmic Ultra-differentiable Class ($W^U_{s,\theta}$): Regularity $e^{s(\ln |j|)^\theta}$ with $\theta > 1$.

Kernel Types:

• Polynomial Decay: $K_k = |k|^{-p}$ (for $k \neq 0$).
• Exponential Decay: $K_k = e^{-|k|^\beta}$ (with $\beta \ge 1$).

Stability Times ($T_r$):
For initial data $u(0)$ in a ball of radius $r$ (excluding a small resonant set), the solution satisfies $\|u(t)\|_s \le 2\|u(0)\|_s$ for time $|t| \le T_r$:

1. Gevrey + Polynomial Kernel: $T_r \sim \exp\left( C \frac{|\ln r|^2}{\ln |\ln r|} \right)$.
2. Gevrey + Exponential Kernel: $T_r \sim \exp\left( C \frac{|\ln r|^2}{\ln |\ln r|} \right)$.
3. Ultra-differentiable ($\theta$) + Polynomial Kernel: $T_r \sim \exp\left( C |\ln r|^{\frac{2\theta}{\theta+1}} \right)$.
4. Ultra-differentiable ($\theta$) + Exponential Kernel: $T_r \sim \exp\left( C |\ln r|^{\frac{2\theta}{\theta+1}} \right)$.

Measure Estimate:
In all cases, the Lebesgue measure of the set of initial data for which stability fails is bounded by:
$$\text{meas}(\text{Resonant Set}) \le r^a \cdot \text{meas}(\text{Ball})$$
where $a < 2/5$. This is a significant improvement over previous bounds.

5. Significance

• Theoretical Advancement: This work bridges a gap between finite-dimensional Nekhoroshev theory and infinite-dimensional PDE stability, specifically addressing the difficult case of internal parameters. It demonstrates that stability can be proven for the "original" equation without artificial parameter modifications.
• Physical Relevance: The non-local Schrödinger equation models various physical phenomena, including self-gravitating quantum matter (Schrödinger-Newton system) and nonlinear optics in non-local media. The results provide rigorous justification for the long-term stability of these systems under small perturbations.
• Methodological Impact: The introduction of the vector field norm for rational normal forms simplifies the technical complexity of handling rational nonlinearities, potentially making these techniques more accessible for future studies of other Hamiltonian PDEs.
• Regularity Frontiers: By successfully handling logarithmic ultra-differentiable regularity, the paper expands the scope of stability theory to function spaces that are physically relevant but mathematically challenging due to their intermediate nature between smooth and analytic.