🔢 mathematics

Boundary Estimates for the Monge-Ampère Equation in the Polygons with Guillemin Boundary Conditions

This paper establishes sharp Euclidean $C^{1,\alpha}$ boundary regularity for the two-dimensional singular Monge-Ampère equation on convex polytopes with Guillemin boundary conditions and Hölder continuous right-hand sides, extending previous results by combining Donaldson's techniques with refined blow-up and localization arguments.

Original authors: Masoud Bayrami-Aminlouee, Reza Seyyedali, Mohammad Talebi

Published 2026-07-28

📖 9 min read🧠 Deep dive

CC BY 4.0

Original authors: Masoud Bayrami-Aminlouee, Reza Seyyedali, Mohammad Talebi

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 you are trying to bake the perfect cake, but the recipe is written in a language of pure geometry. In the world of mathematics, specifically a field called complex geometry, scientists are obsessed with finding "perfect" shapes for spaces, known as manifolds. Think of these shapes as the invisible stage upon which the universe's forces might play out. For decades, mathematicians have been trying to figure out how to build these stages so they are perfectly smooth and balanced. A key tool in this quest is a very tricky equation called the Monge-Ampère equation. You can think of this equation as a strict architect's blueprint that tells you exactly how to curve a surface.

However, there's a catch. Real-world shapes aren't always perfect circles or smooth ovals; sometimes they are polygons, like a stop sign or a slice of pizza, with sharp corners and straight edges. When you try to use the architect's blueprint on these jagged shapes, the math gets messy. The instructions start to scream "infinity!" right at the edges and corners. For a long time, mathematicians could only solve this puzzle if the instructions (the "right-hand side" of the equation) were perfectly smooth, like silk. But what if the instructions are a bit rough, like sandpaper? That's the big question this paper tackles: Can we still build a perfect, smooth shape if the blueprint itself is a little bit bumpy?

The authors of this paper, Masoud Bayrami, Reza Seyyedali, and Mohammad Talebi, have successfully cracked this problem for two-dimensional shapes (polygons). They proved that even if the instructions are only "Hölder continuous"—a fancy way of saying they are rough but not completely chaotic—you can still find a solution that is smooth enough to be useful. They showed that the "roughness" of the instructions doesn't ruin the final shape; it just changes how the shape behaves near the edges. Specifically, they proved that the solution is smooth up to the boundary, including the tricky corners, with a specific level of precision. They didn't just guess this; they built a rigorous mathematical proof using a mix of old tricks and new ideas to show that the solution exists and behaves exactly as they predicted.

The Story of the Jagged Blueprint

Let's dive into the adventure. Imagine you are an architect tasked with designing a roof for a building shaped like a polygon (a shape with straight sides and sharp corners, like a pentagon or a hexagon). Your goal is to make the roof curve in a very specific way so that it holds up under pressure. The rules for how the roof should curve are given by the Monge-Ampère equation.

Now, usually, architects love smooth instructions. If the instructions say, "Curve the roof gently here," and that instruction is perfectly smooth, the math is easy. But in this paper, the authors are dealing with a situation where the instructions are a bit rough. Imagine the instructions are written on a piece of sandpaper. The paper says, "The roughness is limited; it's not jagged, just a little gritty." In math terms, the "right-hand side" of the equation (the part that tells the roof how to curve) is only Hölder continuous. This means it's not perfectly smooth, but it's not totally broken either.

The big problem is that the building itself has corners. In the world of smooth shapes, corners are easy to ignore. But in a polygon, the corners are where the magic (and the trouble) happens. When you try to solve the equation near a corner, the math usually goes wild. The slope of the roof might try to shoot up to infinity, or the curvature might explode. Previous mathematicians, like Rubin and Huang, had solved this puzzle, but only when the instructions were perfectly smooth (like silk). They showed that if the instructions were smooth, the roof would be smooth right up to the edge, even at the corners.

But what if the instructions were rough? That's the gap this paper fills. The authors wanted to know: If the instructions are only "sandpaper smooth," can we still guarantee that the roof is built correctly?

The Magic of the "Guillemin" Boundary

To solve this, the authors use a special rule called the Guillemin boundary condition. Think of this as a special "glue" that holds the roof to the walls of the building. Normally, if you try to glue a roof to a jagged wall, it might peel off or crack. But the Guillemin condition is a very specific type of glue that knows exactly how to handle the jaggedness. It says, "Don't worry about the sharp corner; just make sure the roof behaves like a logarithmic curve near the edge."

The authors realized that the "roughness" of the instructions (the sandpaper) interacts with this special glue in a predictable way. They proved that even with the rough instructions, the roof (the solution) stays smooth enough. They didn't just say "it works"; they proved it with a level of precision called $C^{1,\alpha}$ .

Let's break down what that means in plain English.

$C^1$ means the roof is smooth enough that you can draw a tangent line (a straight line that just touches the curve) at every point, and that line doesn't jump around.
$\alpha$ is a number that measures how "smooth" the slope is. If $\alpha$ is close to 1, the slope changes very gently. If it's smaller, the slope might wiggle a bit more.

The authors proved that the slope of the roof is $C^{1,\alpha}$ smooth. This is a big deal because it's the best possible smoothness you can get when the instructions are rough. They also showed that the curvature (how much the roof bends) behaves in a very specific way near the corners. It doesn't explode into chaos; instead, it grows at a predictable rate, like a balloon inflating at a steady pace.

The Detective Work: Blowing Up and Zooming In

How did they prove this? They used a clever detective technique called localization and blow-up. Imagine you are looking at a map of a city. If you zoom out too far, the streets look like a mess. If you zoom in too close, you just see a single brick. The authors decided to zoom in on the problem areas—the edges and the corners.

They imagined taking a tiny piece of the roof near a corner and stretching it out, like blowing up a balloon. This is the "blow-up" part. When they stretched it, they expected to see a mess because the instructions were rough. But instead, they found that the stretched-out shape looked surprisingly regular. It was like taking a crumpled piece of paper and ironing it out; the wrinkles smoothed out when you looked at them from the right distance.

They used a method inspired by a mathematician named Donaldson, who studied similar problems. They created a "model" roof (a perfect, idealized version) and compared their real, rough roof to it. They proved that the real roof never strays too far from the model. Even though the instructions were rough, the roof stayed close to the perfect shape, just with a little bit of extra wiggle room.

The Two-Step Victory

The paper's proof happens in two main stages, like climbing a mountain with two camps.

Camp 1: The Rough Climb ( $C^{1,\alpha/2}$ )
First, the authors climbed to a lower camp. They proved that the roof is at least $C^{1,\alpha/2}$ smooth. This means the slope is smooth, but maybe a little bit wobbly. The exponent $\alpha/2$ is half of the best possible smoothness. This was a crucial step because it showed that the solution exists and is well-behaved, even if it's not perfectly smooth yet. They did this by carefully analyzing the edges and the corners separately, using special "barrier" functions (imaginary walls) to keep the solution from running away.

Camp 2: The Summit ( $C^{1,\alpha}$ )
Once they were at the lower camp, they used a powerful tool from the world of complex geometry to climb to the summit. They realized that the problem could be translated into a different language (complex coordinates) where the math was easier. In this new language, the rough instructions looked smoother. By solving the problem there and translating it back, they were able to upgrade their result from "half-smooth" to "fully smooth" ( $C^{1,\alpha}$ ).

They also proved that the curvature of the roof near the boundary doesn't just behave nicely; it follows a precise rule. If you get very close to the edge, the curvature grows, but it grows at a rate of $dist(x, \partial P)^{-1+\alpha}$ . This is a fancy way of saying: "The closer you get to the edge, the more the roof bends, but it bends in a way that we can predict and control."

Why This Matters

Why should a curious teenager care about a math paper about polygon roofs? Because this isn't just about roofs. It's about understanding how nature handles roughness. In physics, the universe is full of shapes with edges and corners. If we want to understand how energy or gravity behaves near a black hole (which has a "singularity" or a sharp point) or near the edge of a crystal, we need to know how equations behave in these rough places.

This paper shows that even when the rules are a bit messy, the universe (or at least the mathematical models of it) still finds a way to be orderly. It proves that we don't need perfect, silk-smooth instructions to build a perfect structure. We can work with sandpaper instructions, as long as we know the right tricks.

The authors didn't just guess this; they provided a rigorous proof. They showed that for any polygon in two dimensions, if the instructions are Hölder continuous, there is a unique solution that is smooth up to the boundary. They also ruled out the idea that the solution might be chaotic or undefined at the corners. They proved that the solution is well-behaved, predictable, and smooth.

In short, this paper is a victory for order over chaos. It tells us that even in the jagged, rough corners of the mathematical world, there is a hidden smoothness waiting to be found, as long as we know how to look. The authors have given us a new set of tools to explore these rough edges, and they've shown us that the math holds up, even when the instructions are a little bit gritty.

Technical Summary: Boundary Estimates for the Monge-Ampère Equation in Polygons with Guillemin Boundary Conditions

Problem Statement
This paper addresses the boundary regularity of solutions to the singular real Monge-Ampère equation on two-dimensional convex polytopes $P \subset \mathbb{R}^2$ , subject to the Guillemin boundary condition. The equation is given by:
$\det D^2 u = \frac{H(x)}{l_1(x) \cdots l_N(x)} \quad \text{in } P,$
where $l_i(x)$ are affine linear functions defining the edges of the polytope, and $H$ is a strictly positive function. The solution $u$ must satisfy the Guillemin boundary condition:
$u - \sum_{i=1}^N l_i \log l_i \in C^\infty(P).$
The primary challenge lies in the singularity of the equation along the boundary $\partial P$ , where the gradient of the solution blows up, and the domain lacks smoothness and uniform convexity. While previous works by Rubin [57] and Huang [43] established existence and regularity results for smooth right-hand sides $H$ , this paper focuses on the case where $H$ is merely Hölder continuous ( $H \in C^{0,\alpha}(P)$ ). The goal is to establish sharp a priori estimates and boundary regularity for the "regular part" of the solution, $v = u - u_0$ , where $u_0 = \sum l_i \log l_i$ .

Methodology
The authors employ a multi-stage approach combining techniques from Donaldson's study of the Abreu equation, Caffarelli's localization theory, and the blow-up arguments of Trudinger and Wang.

Uniform Ellipticity via Donaldson's Quantities:
The analysis begins by establishing a uniform equivalence between the Hessian of the solution $D^2 u$ and the Hessian of the model potential $D^2 u_0$ . To handle the blow-up of derivatives at the boundary, the authors utilize Donaldson's normalization technique. They define specific quantities (denoted as $D$ near edges and $E$ near vertices) that measure the deviation of the solution from the model. By assuming these quantities are unbounded and performing a rescaling (blow-up) argument, they derive a contradiction. This relies on showing that the rescaled solutions converge to a nowhere strictly convex function, which violates the boundary conditions derived from the Guillemin structure.
Edge Regularity (Section 3):
Near the edges (away from vertices), the authors use a partial Legendre transform to convert the problem into a quasilinear equation. They establish that the quantity $D$ is uniformly bounded, which allows the application of Caffarelli's interior regularity theory to the rescaled solutions. This yields $C^{2,\alpha}$ estimates for the rescaled potential, translating back to the desired Hessian bounds for $u$ .
Vertex Regularity (Sections 4 & 5):
The analysis near vertices is more delicate due to the coalescence of singularities.
- Diagonal Estimates: The authors first prove boundedness of the quantity $E$ along the diagonal near the vertex.
- Off-Diagonal Estimates: They extend this to points where coordinates are not comparable ( $p_1 \gg p_2$ ) by adapting the barrier construction of Huang. A key technical innovation is modifying Huang's barriers to work for Hölder continuous $H$ by approximating general $H$ with separated functions of the form $F(x_1)G(x_2)$ near the vertex.
- Sharp Regularity: Using the approximation scheme, they prove that the regular part $v$ satisfies a pointwise bound $|v(x)| \le C|x|^{1+\alpha}$ at the vertex.
Complex Geometric Bootstrapping (Section 7):
To upgrade the regularity from $C^{1,\alpha/2}$ to the sharp $C^{1,\alpha}$ , the authors leverage the connection between the real Monge-Ampère equation on toric manifolds and the complex Monge-Ampère equation. By passing to toric complex coordinates (Legendre transform), they show that the singular terms cancel out, leaving a complex Monge-Ampère equation with a $C^{0,\alpha}$ right-hand side and a uniformly elliptic metric. Applying local $C^{2,\alpha}$ theory for the complex equation (Tosatti-Wang-Weinkove-Yang), they obtain regularity for the complex potential. This regularity is then transferred back to the symplectic side to improve the Hölder exponent of the gradient of $v$ from $\alpha/2$ to $\alpha$ .

Key Contributions and Results

Extension to Hölder Continuous Data: The paper extends the existence and regularity results of Rubin and Huang to the case where the right-hand side $H$ is only Hölder continuous ( $C^{0,\alpha}$ ), rather than smooth.
Sharp A Priori Estimates: The main result (Theorem 1.0.3) establishes the following uniform matrix inequality in the sense of symmetric matrices:
$C^{-1} D^2 u_0 \le D^2 u \le C D^2 u_0 \quad \text{in } P,$
where $C$ depends only on the data of the problem (polytope geometry, bounds on $H$ , and $\|u\|_{C^0}$ ).
Boundary Regularity: The authors prove that the regular part $v = u - \sum l_i \log l_i$ belongs to $C^{1,\alpha}(P)$ . Specifically:
$\|v\|_{C^{1,\alpha}(P)} \le C.$
Furthermore, they establish the optimal pointwise blow-up rate for the second derivatives of $v$ near the boundary:
$\text{dist}(x, \partial P)^{-\alpha+1} |D^2 v(x)| \le C.$
Existence of Solutions: As a consequence of the a priori estimates, the paper proves the existence of a unique classical solution to the Dirichlet problem with prescribed vertex values, even when $H$ is only Hölder continuous (Theorem 1.0.4).

Significance
The paper claims that the primary difficulty in this problem is not merely the improvement of the Hölder exponent from $\alpha/2$ to $\alpha$ , but rather the establishment of the initial boundary regularity (the weak Guillemin condition) in the absence of smooth data. The work demonstrates that the specific geometric structure of the polytope and the Guillemin boundary condition are essential for solvability; a naive formulation on smooth domains with similar singularities would not admit solutions. By combining Donaldson's geometric insights with the analytic techniques of Trudinger-Wang and complex Monge-Ampère theory, the authors provide a robust framework for analyzing singular Monge-Ampère equations on non-smooth domains with degenerate boundary conditions. This result is presented as a necessary step toward understanding the existence of constant scalar curvature Kähler (cscK) metrics and extremal metrics on toric manifolds with less regular data.

The Story of the Jagged Blueprint

The Magic of the "Guillemin" Boundary

The Detective Work: Blowing Up and Zooming In

The Two-Step Victory

Why This Matters

More like this