A Circular Cantor Construction Without Dimension Drop for Harmonic Measure
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.
Ada: We're looking at the paper: "A Circular Cantor Construction Without Dimension Drop for Harmonic Measure".
Emmy: Here's the gist of the paper. For every 0 < s < 1 we construct a compact s-Ahlfors regular set K ⊂ C for which harmonic measure of C - K with pole at infinity is comparable to Hs K<ref:2610.15759pg3.
Ada: In particular, there is no dimension drop for any dimension s < 1.<ref:2610.15759#pg4> The main result Theorem 1.1 states that for every 0 < s < 1 there exists a compact s-Ahlfors regular set K ⊂ C such that, with Ω = C - K, ω∞ ≈ Hs K<ref:2610.15759#pg4> In particular, dimH ω∞ = s.<ref:2610.15759#pg4> The construction idea The construction follows a potential-theoretic approach where an N-adic family of points is constructed based on the roots of unity, and small polynomial deformations are introduced to correct the mean potential<ref:2610.15759#pg4> The reference positions are the roots of unity ζj = e 2πij/N, which have an exactly constant discrete logarithmic interaction<ref:2610.15759#pg5> To compensate for the variation in the field generated by other families of points, a polynomial VQ(ζ) = ζ + X M r=1 dQ,rζ r+1 is introduced to correct the first Fourier mode<ref:2610.15759#pg4> The inductive process The construction proceeds inductively through generations, where the goal is to find a number cn such that gn(x) - cn ≤ AN-n, x ∈ Kn<ref:2610.15759#pg3> The key observation is that gn(xQj) - cn = gn−1(Q) + N −nF Q − cn + N −n FQ(j) - F Q<ref:2610.15759#pg4> The three terms in FQ play different roles, with the exterior term ∆2 providing the leading first-mode error of size O(Nτ)<ref:2610.15759#pg4> The role of corrections The scalar parameter correction is chosen to remove the reference-mean error exactly, as HQ - cn + N −n (N − 1)tQ = 0<ref:2610.15759#pg3> The shape parameter correction dQ is then chosen to cancel the leading first-mode response of ∆1, resulting in dQ ≤ Cτ<ref:2610.15759#pg4> This ensures that the remaining nonconstant part of the potential is of size O(Nτ 2)<ref:2610.15759#pg4> The final limit The passage to the limiting equilibrium measure shows that µn converges weakly to the natural Bernoulli measure µ on K<ref:2610.15759#pg4> The geometric argument shows that for fixed M and s < (M + 1)/(M + 2), one can iterate the construction through all generations<ref:2610.15759#pg4> This leads to the conclusion that K and µ are s-Ahlfors regular, and µ is the logarithmic equilibrium measure of K<ref:2610.15759#pg4> The role of polynomial corrections For arbitrary order M, a polynomial deformation VQ(ζ) = ζ + X M r=1 dQ,rζ r+1 is used, and the analysis shows that PN,k(dQ,1,..., dQ,k−1) ≤ CMNτ k<ref:2610.15759#pg3> The principal part of the k-th Fourier mode depends linearly on dQ,k and otherwise only on dQ,1,..., dQ,k−1<ref:2610.15759#pg4> The final conclusion For fixed M and s < 1, the construction is iterated through all generations by choosing N sufficiently large so that all smallness and admissibility conditions hold<ref:2610.15759#pg4> The argument of Section 5 also shows that (cn) is a Cauchy sequence, which implies that cn converges to some c∞ ∈ R<ref:2610.15759#pg4> This proves the final assertion and completes the proof<ref:2610.15759#pg4> References
Bat96: Athanassios Batakis.
Emmy: Harmonic measure of some Cantor type sets.
Ada: Ann.
Paper summary: Ada: So we were talking about the core idea, which is that they are constructing these compact sets K in the complex plane for any dimension s between zero and one, where the harmonic measure of their complement is comparable to their Hausdorff measure.
Emmy: The central thesis is that there's no dimension drop for any s less than one when you use this specific circular construction method.
Ada: That means if you start with a boundary that has dimension s less than one, you don't lose the dimension of its harmonic measure when considering the complement with a pole at infinity.
Kolya: From a probabilistic standpoint, what this means is that the measure we're looking at isn't suffering from some kind of dimensional collapse; it stays in that expected dimension s.
Emmy: The construction relies on an N-adic family based on roots of unity, and they introduce small polynomial deformations to fix the logarithmic potential issues arising from other families of points.
Ada: They use a specific correction term called V Q over zeta plus X M r=one d Q over r plus one, where the magnitude of d Q is controlled by M times tau over r, and tau is defined as N minus one over s <ref:2610.15759#pg1>.
Emmy: This polynomial deformation has the job of successively canceling the first M nonconstant Fourier modes of the local potential, which keeps the remaining error small enough.
Ada: They also introduce a scalar dilation to correct for a mean potential error, and they show that this means only the first M modes are affected by these corrections.
Kolya: The paper mentions that this approach works for arbitrary order M, meaning you can use corrections of any finite order without losing the property.
Emmy: And eventually, when you take the limit over all generations, the resulting measure converges weakly to the natural Bernoulli measure on K.
Ada: So in summary, they've shown a way to build these sets and prove that they maintain their regularity and dimension across this entire range of s less than one.
Conclusion: Ada: So, looking at "A Circular Cantor Construction Without Dimension Drop for Harmonic Measure," it really boils down to how you can construct fractal sets in a way that their harmonic measure behaves nicely even when you are deep into the fractional dimension range.
Emmy: The authors, Yingying Cai, Cole Jeznach, and Xavier Tolsa, found a specific way to handle the potential variation inherent in their construction by using polynomial corrections based on roots of unity.
Ada: What they accomplished is showing that for any s less than one, you can construct an s-Ahlfors regular set K such that its harmonic measure's dimension is exactly s.
Kolya: For someone who only cares about the outcome, this means there isn't a sudden drop in dimension when you look at these sets in this context; it holds for all s less than one.
Emmy: It implies that the structure of these sets, even when they are built recursively and involve complex deformations, is robust enough to preserve its regularity properties under certain conditions.
Ada: The implication is that we have a method to generate these sets where the dimension of the harmonic measure is exactly what we want it to be, which is s.
Kolya: It's a constructive result, so it gives us an explicit way to build these objects rather than just proving they exist abstractly.
Emmy: So this work shows that even with complicated recursive constructions involving deformations, the underlying geometric structure can maintain its desired dimension properties as long as those deformations are controlled in a specific way.
Department de Matematiques, University Autonoma de Barcelona · ICREA, Barcelona, Department of Mathematics, University Autonoma de Barcelona
math.AP
Submitted: 2026-10-11
Updated: 2026-11-11
Comments: 33 pages. Revised manuscript and title
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 92/100
The gist: For every 0 < s < 1 we construct a compact s-Ahlfors regular set K ⊂ C for which harmonic measure of C - K with pole at infinity is comparable to Hs K<ref:2610.15759pg3.
Key concepts
- s-Ahlfors regular set
- This refers to a geometric object K in complex space where its boundary properties are well-behaved concerning a specific dimension s. In essence, it means that the measure of small neighborhoods around points on K scales predictably with respect to the dimension s, which is crucial for understanding how harmonic functions behave near this set.
- Harmonic Measure ($ ext{H}_ u$)
- Harmonic measure measures how much of a region is 'seen' by harmonic functions originating from a specific point (like infinity). In this paper, it quantifies the size of the set C - K relative to its boundary properties. The main result shows that for any s < 1, this measure is comparable to $s^H |K|$.
- Potential-Theoretic Construction
- This is a method of building complex sets by using potential theory, which involves constructing points based on roots of unity and then making small polynomial adjustments. The goal is to correct the mean potential generated by these point families to ensure the resulting set has the desired regularity properties.
- Dimension Drop
- A dimension drop refers to a situation where a geometric property, such as Hausdorff dimension or harmonic measure dimension, suddenly decreases when moving from one configuration or set to another. This paper proves that for this specific construction, there is no such drop; the dimension remains fixed at s for all s in (0, 1).
Terminology
Summary
For every 0 < s < 1 we construct a compact s-Ahlfors regular set K ⊂ C for which harmonic measure of C - K with pole at infinity is comparable to Hs K<ref:2610. p. 015759pg3. In particular, there is no dimension drop for any dimension s < 1. p. 0<ref:2610.15759#pg4>
The main result
Theorem 1.1 states that for every 0 < s < 1 there exists a compact s-Ahlfors regular set K ⊂ C such that, with Ω = C - K, ω∞ ≈ Hs K<ref:2610. p. 115759#pg4> In particular, dimH ω∞ = s. p. 1<ref:2610.15759#pg4>
The construction idea
The construction follows a potential-theoretic approach where an N-adic family of points is constructed based on the roots of unity, and small polynomial deformations are introduced to correct the mean potential<ref:2610.15759#pg4> The reference positions are the roots of unity ζj = e 2πij/N, which have an exactly constant discrete logarithmic interaction<ref:2610. p. 215759#pg5> To compensate for the variation in the field generated by other families of points, a polynomial VQ(ζ) = ζ + X M r=1 dQ,rζ r+1 is introduced to correct the first Fourier mode<ref:2610. p. 215759#pg4>
The inductive process
The construction proceeds inductively through generations, where the goal is to find a number cn such that gn(x) - cn ≤ AN-n, x ∈ Kn<ref:2610.15759#pg3> The key observation is that gn(xQj) - cn = gn−1(Q) + N −nF Q − cn + N −n FQ(j) - F Q<ref:2610. p. 915759#pg4> The three terms in FQ play different roles, with the exterior term ∆2 providing the leading first-mode error of size O(Nτ)<ref:2610. p. 915759#pg4>
The role of corrections
The scalar parameter correction is chosen to remove the reference-mean error exactly, as HQ - cn + N −n (N − 1)tQ = 0<ref:2610.15759#pg3> The shape parameter correction dQ is then chosen to cancel the leading first-mode response of ∆1, resulting in dQ ≤ Cτ<ref:2610. p. 615759#pg4> This ensures that the remaining nonconstant part of the potential is of size O(Nτ 2)<ref:2610. p. 715759#pg4>
The final limit
The passage to the limiting equilibrium measure shows that µn converges weakly to the natural Bernoulli measure µ on K<ref:2610.15759#pg4> The geometric argument shows that for fixed M and s < (M + 1)/(M + 2), one can iterate the construction through all generations<ref:2610. p. 315759#pg4> This leads to the conclusion that K and µ are s-Ahlfors regular, and µ is the logarithmic equilibrium measure of K<ref:2610. p. 415759#pg4>
The role of polynomial corrections
For arbitrary order M, a polynomial deformation VQ(ζ) = ζ + X M r=1 dQ,rζ r+1 is used, and the analysis shows that PN,k(dQ,1,..., dQ,k−1) ≤ CMNτ k<ref:2610. p. 015759#pg3> The principal part of the k-th Fourier mode depends linearly on dQ,k and otherwise only on dQ,1,. p. 7.., dQ,k−1<ref:2610. p. 015759#pg4>
The final conclusion
For fixed M and s < 1, the construction is iterated through all generations by choosing N sufficiently large so that all smallness and admissibility conditions hold<ref:2610.15759#pg4> The argument of Section 5 also shows that (cn) is a Cauchy sequence, which implies that cn converges to some c∞ ∈ R<ref:2610.15759#pg4> This proves the final assertion and completes the proof<ref:2610.15759#pg4>
References
[Bat96] Athanassios Batakis. Harmonic measure of some Cantor type sets. p. 1 Ann. Acad. Sci. Fenn. Math., 21(2):255–270, 1996<ref:2610. p. 015759#pg4>
[Bou87] Jean Bourgain. On the Hausdorff dimension of harmonic measure in higher dimension. p. 0 Invent. Math., 87(3):477–483, 1987<ref:2610. p. 015759#pg3>
[Wol95] Thomas H. Wolff. Counterexamples with harmonic gradients in R 3. p. 0 In Essays on Fourier Analysis in Honor of Elias M. p. 2 Stein, Princeton Math. p. 0 Ser., vol. 42, pp. 321–384, Princeton Univ Press, 1995<ref:2610. p. 015759#pg3>
[DJJ23] Guy David, Cole Jeznach, and Antoine Julia. Cantor sets with absolutely continuous harmonic measure. p. 1 J. Ec. polytech. Math., 10:1277–1298, 2023<ref:2610. p. 015759#pg4>
[Tol24] Xavier Tolsa. The dimension of harmonic measure on some AD-regular flat sets of fractional dimension. p. 1 Int. Math Res Not IMRN, (8):6579–6605, 2024<ref:2610. p. 015759#pg4>
[JW88] Peter W. Jones and Thomas H. p. 1 Wolff. Hausdorff dimension of harmonic measures in the plane. p. 1 Acta Math., 161(1-2):131–144, 1988<ref:2610. p. 015759#pg3>
[Azz20] Jonas Azzam. Dimension drop for harmonic measure on Ahlfors regular boundaries. p. 0 p. 0 Potential Anal. p. 0, 53(3):1025–1041, 2020<ref:2610. p. 015759#pg5>
[BZ15] Athanasios Batakis and Anna Zdunik. Hausdorff and harmonic measures on non-homogeneous Cantor sets. p. 1 Ann. Acad. Sci. Fenn. Math., 40(1):279–303, 2015<ref:2610. p. 015759#pg4>
[P26] A. Pathak. On the dimension drop for harmonic measure on uniformly non-flat Ahlfors-David regular boundaries. Preprint, arXiv:2601. p. 016167 (2026)<ref:2610. p. 015759#pg4>
[Vol22] Alexander Volberg. One phase problem for two positive harmonic functions: below the codimension 1 threshold. p. 1 Preprint, arXiv:2205. p. 003687, 2022<ref:2610. p. 015759#pg5>
[CJT26] Yingying Cai, Cole Jeznach, and Xavier Tolsa. Failure of dimension drop for harmonic measure outside Ahlfors regular flat boundaries. p. 0 Preprint, 2026<ref:2610. p. 015759#pg4>
[CT26] Yingying Cai and Xavier Tolsa. Dimension drop for harmonic measure on Ahlfors regular boundaries. Preprint, arXiv:2604. p. 021047, 2026<ref:2610. p. 0
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Polynomial correction of arbitrary order capability: The system can construct compact s-Ahlfors regular sets K for any dimension s less than 1, by using polynomial deformations of arbitrary finite order to make the logarithmic potential constant in the limit.
-
Full range coverage for non-flat boundaries: The improved system can resolve the phenomenon where Ahlfors regularity alone does not force dimension drop below dimension one, extending examples to every 0 < s < 1 in the non-flat planar setting.
-
Precise control over harmonic measure estimates: The system can perform detailed analysis of discrete potentials by tracking three error terms (Delta 1, Delta 2, and Delta 3) and showing that the mean is close to a target level cn with an error bounded by CN −nN(C tau) N/2.
-
Adaptive parameter selection: The AI can adapt shape parameters dQ,r inductively in a one-pass triangular correction process to ensure that the first Fourier mode of the potential vanishes, leading to a bound of ||FQ - F Q||infinity <= CMNtau M+1.
-
Convergence to equilibrium measure: The system can demonstrate that the sequence of measures µn converges weakly to the natural Bernoulli measure µ on K, and that this measure is precisely the logarithmic equilibrium measure of K with respect to its s-Ahlfors regularity.
Sources
- On the dimension drop for harmonic measure on uniformly non-flat Ahlfors-David regular boundaries
- One phase problem for two positive harmonic function: below the codimension $1$ threshold
- Dimension Drop for Harmonic Measure on Ahlfors Regular Boundaries
Related papers
- Regularity of critical immersions for scale-invariant curvature energies in even dimensions
- C 1, alpha Regularity of W 2,2 Q-Valued Maps with Small Laplacian
- Cumulative angular cancellation in Maz'ya's endpoint inequalities
- Strictly positive optimal Hardy weights on manifolds and graphs via Green potential
- A fast dynamo with zero topological entropy
- A counterexample to fixed-profile soliton resolution for the three-dimensional quintic wave equation