🔢 mathematics

Singular extension of critical Sobolev mappings with values into complete Riemannian manifolds

This paper extends singular extension results for critical nonlinear Sobolev mappings into compact Riemannian manifolds to the case of complete non-compact target manifolds that admit isometric embeddings with positive reach, establishing exponential weak-type Sobolev-Marcinkiewicz estimates for the extended maps.

Original authors: Federico Luigi Dipasquale

Published 2026-06-16

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Federico Luigi Dipasquale

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 have a crumpled piece of paper (representing a complex shape or "manifold") and you want to smooth it out onto a flat table without tearing it or stretching it too much. In mathematics, this is called an extension problem: you have a map defined on the edge of a shape (like the surface of a sphere), and you want to fill in the inside with a smooth, continuous map that connects everything nicely.

This paper is about solving a specific, very tricky version of this puzzle. Here is the breakdown in plain English:

The Main Problem: The "Tightrope" Walk

Mathematicians have known how to do this "smoothing" for shapes that are small and closed (like a perfect sphere or a donut). But what if your shape is infinite? What if it's an endless plane or a shape that stretches out forever, like a long, winding road that never ends?

The difficulty is that when things get infinite, they can get "too loose" or "too wild." If you try to smooth out a map on an infinite shape, the math might break, and the map could tear or become infinitely jagged.

The Key Ingredient: The "Safety Net" (Reach)

The author introduces a concept called "reach." Think of reach as a safety buffer zone.

Imagine your shape is a wire sculpture floating in a room.
If the sculpture is very twisted, the "safety zone" around it might be tiny because if you get too close, you might accidentally touch two different parts of the wire at once.
If the sculpture is smooth and well-behaved, you can have a thick, clear "safety bubble" around it where you can always tell exactly which part of the wire is closest to you.

The paper says: If your infinite shape has a guaranteed "safety bubble" (positive reach), we can solve the smoothing problem.

The Breakthrough: Using a New Rule

The author uses a recent discovery by another mathematician (A. Petrunin) to figure out which infinite shapes have this "safety bubble."

The Rule: If an infinite shape doesn't grow too fast (it grows like a polynomial, not an explosion) and doesn't have weird sharp corners or infinite curvature, it has a safety bubble.
The Result: This allows the author to take a famous mathematical proof that worked only for small, closed shapes and extend it to these specific, well-behaved infinite shapes.

The "Magic" Estimate

The paper proves that when you smooth out the map on these infinite shapes, the result isn't perfect everywhere. There might be a few tiny spots where the map is a bit rough (singularities). However, the author proves that:

These rough spots are rare (there are only a finite number of them).
The "roughness" is controlled by a specific mathematical formula.
Even if the shape is infinite, the "badness" of the map is kept in check, growing only in a predictable, exponential way.

Real-World Examples (The "What Works" List)

The paper lists specific types of infinite shapes where this new math works:

Warped Cylinders: Imagine a tube where the width changes smoothly as you go down the length, but it doesn't get infinitely thin or thick too fast.
Universal Coverings: Think of a video game map that loops. If you "unroll" it so it never loops back on itself (making it infinite), and the original game world had a simple structure, this new math works on the unrolled version.
Conical Shapes: Shapes that look like a cone stretching out to infinity.

What Doesn't Work (The "Warning Signs")

The paper also points out shapes where this math fails:

Hyperbolic Space: Imagine a surface that curves away from itself so aggressively that its volume explodes exponentially (like a coral reef that gets bigger faster than you can count). This shape is too "wild" to have a safety bubble, so the math breaks down.
Negatively Curved Universes: If a shape curves away from itself in a way that makes its "loops" grow exponentially, it's too chaotic for this method.

Summary

In short, this paper is a bridge. It takes a powerful mathematical tool that was previously stuck on small, closed islands and builds a bridge to specific types of infinite continents. It tells us exactly which infinite continents are "safe" to cross (those with a "reach" or safety buffer) and guarantees that if we cross them, the path we build will be smooth enough to be useful, with only a few manageable bumps along the way.

Technical Summary: Singular Extension of Critical Sobolev Mappings into Complete Riemannian Manifolds

Problem Statement
The paper addresses the problem of extending maps defined on the boundary of a domain into a target Riemannian manifold $N$ that are not necessarily compact. Specifically, it considers maps $u$ belonging to the critical, nonlinear Sobolev space $W^{m/(m+1), m+1}$ , where the domain is either the sphere $S^m$ , the Euclidean plane $\mathbb{R}^m$ , or the boundary of the Poincaré ball model of hyperbolic space. The goal is to extend these maps to the interior (the ball $B^{m+1}_1$ , the half-space $\mathbb{R}^{m+1}_+$ , or the hyperbolic space $H^{m+1}$ ) while controlling the energy of the extension.

Previous results by Bulanyi and Van Schaftingen [5] established such extensions for compact target manifolds $N$ , yielding an exponential weak-type Marcinkiewicz-Sobolev estimate. However, the compactness of $N$ was a restrictive assumption. This paper investigates whether these extension results can be generalized to complete, non-compact Riemannian manifolds.

Methodology and Theoretical Framework
The author's approach relies on two main theoretical pillars:

Re-examination of the Proof in [5]: A careful analysis of the proofs in Bulanyi and Van Schaftingen [5] reveals that the compactness of $N$ is utilized for only two specific purposes:
- To ensure the image of the Nash isometric embedding has a positive reach, guaranteeing the existence of a well-defined, smooth nearest-point projection.
- To ensure the "gap potential" (an integral over pairs of points with distance $\ge \delta$ ) controls the truncated Gagliardo energy.
  The author argues that if $N$ admits an isometric Euclidean embedding with positive reach, and if the range of the map $u$ has a bounded diameter, the compactness assumption can be relaxed.
Petrunin's Criterion for Tubed Embeddings: To identify non-compact manifolds satisfying the "positive reach" condition, the paper utilizes a recent sufficient criterion by A. Petrunin [13]. Petrunin proved that a complete, smooth, connected Riemannian manifold with 1-bounded geometry (positive injectivity radius and bounded first covariant derivative of the curvature tensor) and uniformly polynomial growth (volume of geodesic balls bounded by a polynomial in the radius) admits an isometric Euclidean embedding with positive reach (a "tubed embedding").

Key Contributions and Results
The paper presents three main theorems (Theorems 1, 2, and 3) which extend the singular extension results of [5] to connected Riemannian manifolds $N$ admitting an isometric Euclidean embedding with positive reach.

Extension Theorems: For a map $u \in W^{m/(m+1), m+1}(\partial \Omega, N)$ , there exists an extension $U \in W^{1,1}(\Omega, N)$ such that the trace of $U$ is $u$ . The extension satisfies an exponential weak-type estimate:
$t^{m+1} \mathcal{L}^{m+1}(\{x \in \Omega : |DU(x)| \ge t\}) \le A \exp\left( B \int \int \frac{d(u(x), u(y))^{m+1}}{|x-y|^{2m}} dx dy \right) \int \int \frac{d(u(x), u(y))^{m+1}}{|x-y|^{2m}} dx dy$
where the constants $A, B, \delta$ depend on the dimension $m$ and the reach of $N$ .
Bounded Diameter Case: If the image of the map has a bounded diameter $L$ , the exponential term can be refined to depend on a "gap potential" (integrating only over pairs where $d(u(x), u(y)) \ge \delta$ ), with the constant $B'$ depending on $L$ and the reach.
Singular Set: The extension $U$ can be chosen to be smooth outside a finite singular set $S$ , the cardinality of which is controlled by the right-hand side of the estimates.
Specific Domains: The results cover:
- The unit ball $B^{m+1}_1$ (Theorem 1).
- The upper half-space $\mathbb{R}^{m+1}_+$ (Theorem 2).
- The hyperbolic space $H^{m+1}$ via the Poincaré ball model (Theorem 3).

Significance and Scope
The paper positions itself as a direct extension of existing work rather than a discovery of new extension mechanisms. Its primary significance lies in:

Generalization: It explicitly identifies that the compactness of the target manifold is not strictly necessary, provided the manifold admits a tubed Euclidean embedding (positive reach) and the map's image is bounded.
Application of Petrunin's Theorem: It leverages Petrunin's criterion to demonstrate that a wide class of non-compact manifolds satisfy the necessary geometric conditions. The paper provides illustrative examples of such manifolds, including:
- Warped products $\mathbb{R} \times_f M$ with bounded warping functions.
- Universal coverings of compact manifolds with polynomially growing fundamental groups (e.g., those with non-negative Ricci curvature or Abelian fundamental groups).
- Asymptotically Euclidean and conical manifolds.
Limitations: The paper acknowledges that Petrunin's criterion is sufficient but not necessary. It notes that manifolds with exponential volume growth (like hyperbolic space itself) or universal coverings of manifolds with exponential fundamental group growth do not admit such embeddings and thus fall outside the scope of these results.

The author emphasizes that the proofs follow the strategy of [5] almost verbatim, with modifications confined to the initial setup regarding the reach and the handling of the gap potential for non-compact targets. The results do not claim to solve the extension problem for all complete non-compact manifolds, but rather for those identifiable via Petrunin's geometric criteria.