Sharp quantitative integral inequalities for general conformally invariant extensions

This paper establishes sharp quantitative integral inequalities for a general family of conformally invariant extension operators and their adjoints by developing a refined analysis of hypergeometric functions, thereby extending previous results to the full admissible parameter range.

The Gist

Imagine you are an architect trying to build the most efficient structure possible. In mathematics, there's a famous rule called the Isoperimetric Inequality. Think of it like this: if you have a fixed amount of fencing (perimeter), the shape that encloses the most land (area) is a perfect circle. Any other shape is "wasteful."

For a long time, mathematicians knew exactly what the "perfect circle" looked like in complex, high-dimensional spaces. But they also wanted to know: What happens if your shape is slightly crooked? How much "wastefulness" (or error) do you introduce if you deviate just a tiny bit from the perfect circle?

This paper, written by Qiaohua Yang and Shihong Zhang, answers that question for a very broad and complex family of mathematical shapes and rules. Here is a breakdown of their work using simple analogies.

1. The Big Picture: The "Perfect Shape" and the "Wobble"

The authors are studying a specific type of mathematical operation called a Conformally Invariant Extension.

• The Analogy: Imagine you have a flat, 2D map of a city (the boundary). You want to build a 3D skyscraper (the extension) based on that map. There are many ways to build this skyscraper, but there is one "Golden Standard" way that is perfectly balanced and efficient.
• The Problem: In the past, mathematicians only knew how to measure the "wobble" (stability) when the skyscraper was built using a very specific, simple rule (like a harmonic extension).
• The Breakthrough: Yang and Zhang figured out how to measure the wobble for every possible rule in this family, not just the simple one. They proved that even when the rules get very complicated, the "Golden Standard" is still the most efficient, and they calculated exactly how much efficiency you lose if you stray from it.

2. The Two Types of "Wobbles" (The Degeneracy)

One of the most interesting findings in the paper is about how the "wobble" behaves. The authors discovered that the behavior changes depending on how "heavy" or "light" the mathematical shape is.

• The Light Shapes (Small $p$): Imagine a wobbly jelly. If you push it slightly, it wobbles back in a very smooth, quadratic way (like a spring). The error grows slowly.
• The Heavy Shapes (Large $p$): Imagine a stiff, heavy rock. If you push it, it doesn't wobble smoothly; it resists differently. The error grows much faster, proportional to the size of the push itself.

The Metaphor: Think of a trampoline.

• If you are light (a child), the trampoline bounces back gently (the math behaves like a square, $p=2$).
• If you are very heavy (an elephant), the trampoline stretches differently, and the math behaves like a higher power ($p$).
• The Discovery: The authors found that for their complex skyscrapers, the "stiffness" of the trampoline can be anything. It's not just 2 or 4; it can be 100, 1000, or any number. This was a surprise because previous theories suggested the stiffness was limited.

3. The "Dual" Problem: Looking from the Inside Out

Mathematics often has a "mirror image" problem.

• The Main Problem: You start with a map and build a skyscraper (Extension).
• The Dual Problem: You start with the skyscraper and try to reconstruct the map (Dual Operator).

Usually, if you solve the first problem, the second one is easy to solve by flipping the equation. However, the authors found a twist here.

• The Twist: In the "Main" problem, the perfect shape is a flat, constant surface (like a calm lake). But in the "Dual" problem, the perfect shape is not flat. It's a curved, complex hill.
• Why it matters: This means you can't just use the same simple tricks to solve the dual problem. The authors had to invent new, sophisticated tools to handle this "curved hill" and prove that it is indeed the most efficient shape, even though it looks complicated.

4. The Secret Weapon: Hypergeometric Functions

To prove all of this, the authors had to dive deep into a specific type of mathematical function called Hypergeometric Functions.

• The Analogy: Think of these functions as a universal remote control for complex shapes. They can describe how a shape changes as you zoom in, zoom out, or twist it.
• The Challenge: These functions are notoriously difficult to analyze. The authors had to write a new "user manual" for them, proving exactly how they behave when the parameters change. They showed that even when these functions get messy or blow up (become infinite), the underlying stability of the system remains intact.

Summary: Why Should You Care?

This paper is like upgrading the blueprint for the universe's most efficient structures.

1. Generality: They didn't just fix one specific case; they solved the problem for an entire universe of possibilities.
2. Precision: They didn't just say "it's stable"; they gave the exact formula for how stable it is, revealing that the "stiffness" of the system can be infinitely variable.
3. New Tools: They developed new mathematical techniques to handle "curved" perfect shapes, which will help other scientists solve similar problems in physics and geometry.

In short, Yang and Zhang took a complex, rigid mathematical puzzle and showed us that the solution is flexible, robust, and works for every scenario imaginable, provided you know how to tune the "dials" correctly.

Technical Summary

Here is a detailed technical summary of the paper "Sharp quantitative integral inequalities for general conformally invariant extensions" by Qiaohua Yang and Shihong Zhang.

1. Problem Statement

The paper addresses the stability of sharp conformally invariant integral inequalities on the unit ball $B^n$ and the upper half-space $\mathbb{R}^n_+$. Specifically, it focuses on a general family of extension operators $E_{\alpha, \beta}$ (and their duals $R_{\alpha, \beta}$) which generalize the classical harmonic extension and the Hardy-Littlewood-Sobolev (HLS) inequalities.

The core problem is to establish quantitative stability inequalities (also known as stability estimates) for these operators. While the sharp constants and extremal functions for these inequalities were recently established by Gluck [22], the question of how close a function must be to an extremal function if the inequality is nearly sharp (i.e., the "deficit" is small) remained open for the general parameter range.

The authors aim to prove that the deficit in the inequality is bounded from below by a distance functional measuring the deviation of the function from the manifold of extremals. A key challenge is that the exponent in this distance functional depends on the parameters of the operator, exhibiting a "degeneracy" phenomenon where the exponent changes from 2 to $p$ depending on the range of $p$.

2. Methodology

The authors employ a sophisticated combination of concentration-compactness, spectral analysis, and hypergeometric function theory.

A. Conformal Invariance and Reformulation

The problem is reformulated on the unit ball $B^n$ using conformal maps. The extension operator $Q_{\alpha, \beta}$ is defined via an integral kernel involving the hypergeometric function. The authors utilize the conformal covariance of these operators to reduce the stability problem to analyzing perturbations around the extremal functions (which are generally non-constant for $\alpha \neq 0, \beta \neq 1$).

B. Spectral Gap Analysis via Hypergeometric Functions

A central technical hurdle is establishing a spectral gap for the linearized operator around the extremal.

• Weighted Operator: Unlike the harmonic case ($\alpha=0, \beta=1$) where the extremal is constant, the general extremal $d_{\alpha, \beta}$ is non-constant and may be unbounded. The authors analyze the weighted operator $d_{\alpha, \beta}^{q-2} Q_{\alpha, \beta}$.
• Spherical Harmonic Expansion: They expand functions in spherical harmonics. The eigenvalues of the linearized operator are expressed in terms of integrals involving hypergeometric functions $F(a, b; c; z)$.
• Monotonicity Lemmas: A critical contribution is the proof of monotonicity properties for these hypergeometric coefficients across different parameter regimes ($\alpha < 1$, $\alpha = 1$, $\alpha > 1$). This allows them to identify the second eigenvalue (the spectral gap) explicitly and prove it is strictly positive.

C. Compactness and Concentration-Compactness

• Compactness: Due to the potential unboundedness of the weight $d_{\alpha, \beta}$ when $\alpha + \beta < 1$, standard compactness arguments fail. The authors prove the compactness of the weighted operator by establishing the decay rate of the expansion coefficients $A_l$ as the degree $l \to \infty$.
• Concentration-Compactness: They utilize a concentration-compactness principle to handle sequences that might lose compactness (e.g., concentrating at a point). This ensures that if the deficit goes to zero, the sequence must converge (up to conformal transformations) to an extremal.

D. Second Variation and Distance Functionals

The proof relies on comparing the second variation of the functional with the distance to the manifold of extremals.

• They use refined inequalities (from Figalli and Zhang) to handle the non-quadratic nature of the $L^p$ norm when $p \neq 2$.
• Dual Operator: For the dual inequality, the minimizer is generally non-constant. The authors adapt the duality argument, noting that the standard Carlen duality framework fails here due to the non-constant minimizer and the specific range of exponents ($q' < 2$).

3. Key Contributions

1. Extension to Full Parameter Range: The results extend the recent work of Frank, Peteranderl, and Read [19] (which was limited to the harmonic extension case) to the full admissible parameter range $(\alpha, \beta)$ satisfying the constraints in (1.3).
2. Sharp Stability Exponents: The paper identifies the optimal exponent for the distance functional in the stability inequality.
   • For the primal operator $Q_{\alpha, \beta}$:
     • If $p \ge 2$ (corresponding to $\alpha \le 1$), the distance is measured in $L^p$ with exponent 2 (quadratic) plus an $L^2$ term.
     • If $1 < p < 2$(corresponding to$\alpha > 1$), the distance is measured in$L^p$ with exponent 2.
     • Correction/Refinement based on Theorem 1.1: The paper actually establishes that the exponent is 2 when $p \ge 2$ (combined with an $L^2$ term) and 2 when $1 < p < 2$(purely$L^2$ distance). Wait, let's re-read Theorem 1.1 carefully.
     • Theorem 1.1 (Primal):
       • Case $p \ge 2$ ($\alpha \le 1$): The deficit is bounded by $\inf \| \dots \|_{L^p}^p + \| \dots \|_{L^2}^2$. The dominant term for small perturbations is the $L^2$ term (exponent 2).
       • Case $1 < p < 2$($\alpha > 1$): The deficit is bounded by$\inf | \dots |_{L^p}^2$. The exponent is 2.
       • Note: The abstract mentions a degeneracy where the exponent equals 2 for $p \in (1, 2)$ and becomes $p$ for $p \ge 2$ in some contexts (like Sobolev), but here the result shows the exponent is 2 in the $L^p$ distance for $p < 2$ and a mix for $p \ge 2$. The paper highlights that for the nonlocal operator, the degeneracy can be arbitrarily large, contrasting with local Sobolev inequalities.
3. Dual Operator Stability: The paper provides the first sharp quantitative stability result for the dual operator $S_{\alpha, \beta}$ in the general setting. It proves that the stability holds with an exponent of 2 in the $L^{q'}$ distance, even though the minimizer is non-constant.
4. Hypergeometric Analysis: The rigorous analysis of the monotonicity of hypergeometric function coefficients provides a new tool for studying spectral gaps in nonlocal operators with weights.

4. Main Results

Theorem 1.1 (Stability for Primal Operator $Q_{\alpha, \beta}$):
For $n \ge 3$ and admissible $(\alpha, \beta)$:

• If $p \ge 2$ ($\alpha \le 1$):
  $$c_{\alpha, \beta}^p - \frac{\|Q_{\alpha, \beta} u\|_{L^q}^p}{\|u\|_{L^p}^p} \ge c_0 \left( \inf_{\Psi, \lambda} \|\lambda |S^{n-1}|^{1/p} u_\Psi - 1\|_{L^p}^p + \|\lambda |S^{n-1}|^{1/p} u_\Psi - 1\|_{L^2}^2 \right)$$
• If $1 < p < 2$($\alpha > 1$):
  $$c_{\alpha, \beta}^p - \frac{\|Q_{\alpha, \beta} u\|_{L^q}^p}{\|u\|_{L^p}^p} \ge c_0 \inf_{\Psi, \lambda} \|\lambda |S^{n-1}|^{1/p} u_\Psi - 1\|_{L^p}^2$$
  The exponents in the distance functionals are shown to be optimal.

Theorem 1.2 (Stability for Dual Operator $S_{\alpha, \beta}$):
For the dual inequality, there exists a constant $c_0 > 0$ such that for all $v \in L^{q'}(B^n)$:
$$c_{\alpha, \beta}^{q'} - \frac{\|S_{\alpha, \beta} v\|_{L^{p'}}^{q'}}{\|v\|_{L^{q'}}^{q'}} \ge c_0 \inf_{\Psi, \lambda} \|\lambda |B_n|^{1/q'} v_\Psi - \tilde{1}\|_{L^{q'}}^2$$
where $\tilde{1}$ is the normalized extremal function (which is generally non-constant). The exponent 2 is optimal.

5. Significance

• Generalization: This work significantly broadens the scope of stability analysis from the specific case of harmonic extensions to a wide class of conformally invariant operators, covering the full range of parameters where the inequalities hold.
• Nonlocal Phenomena: It demonstrates that for nonlocal operators, the stability exponent can behave differently than in local Sobolev inequalities. The paper confirms that the "degeneracy" (where the exponent jumps from 2 to $p$) observed in local problems has a distinct manifestation in nonlocal settings, often stabilizing at exponent 2 even for large $p$.
• Methodological Advance: The development of spectral gap lemmas using hypergeometric functions for weighted, non-constant extremals provides a powerful new framework for analyzing stability in geometric inequalities involving nonlocal operators.
• Dual Stability: By resolving the stability for the dual operator with non-constant minimizers, the paper fills a gap in the literature where previous duality arguments (like those of Carlen) were insufficient.

In summary, Yang and Zhang provide a comprehensive and sharp quantitative stability theory for a broad class of conformally invariant extension inequalities, utilizing deep analysis of hypergeometric functions to overcome the challenges posed by non-constant extremals and nonlocal operators.