Riesz transforms and rectifiability in higher codimension

1 · The essentials

This result connects a test based on averaging with the shape of a distribution of mass. It concerns mass spread through a space with at least four dimensions, but behaving as though it occupies fewer dimensions. A key assumption is that the amount of mass in a ball stays between fixed multiples of the ball’s radius raised to the appropriate power. The test combines contributions from the surrounding mass while leaving out points closer than a chosen distance. If its size remains controlled by the same bound for every positive choice of that distance, the result guarantees substantial pieces of the mass with controlled geometric descriptions. The geometric bounds depend only on the dimensions, the mass-growth bounds, and the bound for the test. The main result has a Lean formalization.

A picture to help (an imperfect image, not to be taken literally)

As an imperfect image, imagine examining a cloud of dots through a detector that ignores everything inside an adjustable nearby circle. The result says that, under a precise rule about how densely the dots are distributed, consistent control of the detector at every exclusion distance forces large portions of the cloud to have controlled shapes.

Why it matters

The manuscript gives a positive answer to the David–Semmes Riesz-transform question in what the source describes as its remaining higher-codimension range. It turns a bound on an averaging operator into a quantitative geometric conclusion. In particular, the conclusion supplies uniform big pieces of Lipschitz images of Euclidean balls, with bounds controlled by the stated input data.

2 · Going further

The precise setting is an $n$ -Ahlfors–David regular Radon measure $μ$ on $R^{d}$ , for integers $d \geq 4$ and $2 \leq n \leq d - 2$ . Ahlfors–David regularity requires the mass of balls centered on the support, at the relevant scales, to be bounded above and below by fixed multiples of $r^{n}$ . The problem is whether boundedness of the $n$ -dimensional Riesz transform forces uniform $n$ -rectifiability in this higher-codimension range.

The operator hypothesis is uniform boundedness from scalar $L^{2} (μ)$ to vector-valued $L^{2} (μ)$ over all positive hard truncations. A hard truncation excludes contributions from points within a chosen positive distance of the evaluation point; the same operator bound must work for every such distance. The conclusion is uniform $n$ -rectifiability, expressed in the abstract as uniform big pieces of Lipschitz images of Euclidean balls. Its quantitative bounds depend only on dimension, the regularity bounds, and the operator bounds.

The source identifies this as a positive answer to the David–Semmes Riesz-transform question in its remaining higher-codimension range, and labels the main manuscript Lean-formalized. The supplied abstract states the implication but does not describe the argument’s mechanism, so no proof strategy can be extracted from it. Only the main manuscript is supplied; there is no information here about companion manuscripts or their relationship to the result.

Useful notions

Measures and mass in balls
Euclidean dimension and codimension
Scalar and vector-valued square-integrable functions
Operator bounds and truncations
Lipschitz maps and geometric regularity

What was known before / what is new

Before

The source identifies the David–Semmes Riesz-transform question as the problem being addressed. It describes the stated higher-codimension range as the remaining range, but gives no account of earlier results or their methods.

New

The main result establishes that uniform Riesz-transform bounds imply uniform rectifiability for the stated Ahlfors–David regular measures when $d \geq 4$ and $2 \leq n \leq d - 2$ . The conclusion includes uniform big pieces of Lipschitz images of Euclidean balls. The geometric bounds depend only on dimension, regularity, and operator bounds.

Limits and assumptions

The theorem requires an $n$ -Ahlfors–David regular Radon measure and a single scalar-to-vector $L^{2} (μ)$ operator bound valid for every positive hard truncation. Its stated dimensional range is $d \geq 4$ and $2 \leq n \leq d - 2$ ; the supplied statement does not address weaker hypotheses or other ranges. The main result has a Lean formalization, but the source provides no finer inventory of formalized auxiliary claims. The supplied abstract gives neither argument details nor a list of questions that remain open.

Mini glossary

Ahlfors–David regularity: A mass-growth condition requiring the measure of balls centered on the support, at the relevant scales, to stay between fixed positive multiples of $r^{n}$ .
Codimension: The difference $d - n$ between the ambient dimension and the dimension assigned to the measure.
Hard truncation: An exclusion of contributions from points within a chosen positive distance when evaluating the transform.
$L^{2} (μ)$: The space of functions whose squared magnitude has finite integral with respect to $μ$ .
Lipschitz map: A map that increases distances by at most a fixed factor.
Uniform rectifiability: The quantitative geometric property expressed here by uniform big pieces of Lipschitz images of Euclidean balls.

Keywords : Riesz transforms · Uniform rectifiability · Higher codimension

AI-generated explanation based on the manuscript. Unofficial. Refer to the manuscript for the authoritative version.

3 · Technical details and sources

Claim made in the manuscript (source text)

Resolves the remaining higher-codimension Riesz-transform rectifiability problem: for d ≥ 4 and $2 \leq n \leq d - 2$ , an n-Ahlfors–David regular Radon measure on ℝ^d is uniformly n-rectifiable whenever its n-dimensional Riesz transform is uniformly L²-bounded over all positive hard truncations. The rectifiability bounds depend only on dimension, regularity and operator bounds.

Verified by Lean: the Lean software checked the part of the reasoning described below, sometimes only part of the manuscript. This does not mean independent mathematicians have validated the result.

Scope of the Lean formalization (source text)

…

lean/docs/081.md

Manuscripts in this family

★

Riesz transforms and uniform rectifiability in higher codimension

24/09/2026

Verified by Lean

We prove that an n-Ahlfors–David regular Radon measure on ℝ^d is uniformly n-rectifiable if its n-dimensional Riesz transform is uniformly bounded from scalar $L^{2} (μ)$ to vector-valued $L^{2} (μ)$ over all positive hard truncations, for integers d ≥ 4 and $2 \leq n \leq d - 2$ . This gives a positive answer to the David–Semmes Riesz-transform question in its remaining higher-codimension range. The conclusion gives uniform big pieces of Lipschitz images of Euclidean balls.

PDF GitHub folder Lean formalization