Conductive homogeneity of locally symmetric polygon-based self-similar sets

Jun Kigami; Yuka Ota

REVIEW 2 major objections 6 minor 16 references

Conductive homogeneity of locally symmetric polygon-based self-similar sets

T0 review · 2 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Locally symmetric polygon-based self-similar sets are $p$-conductively homogeneous for every $p$ above the Ahlfors regular conformal dimension, so they carry Sobolev-type spaces and, at $p=2$, Brownian motion with a continuous heat kernel.

desk verdict Real new examples and a useful reduction to finite combinatorics, but the proof of the second backbone theorem leaves an unhandled case that the later sufficient conditions depend on. read the letter →

arxiv 2505.23057 v1 pith:NYHANMD2 submitted 2025-05-29 math.MG

Jun Kigami , Yuka Ota This is my paper

classification math.MG MSC 28A8031C2560J65

keywords conductivehomogeneityself-similarsetsregularpolygontilingslocalsymmetryp-energyDirichletformsBrownianmotiononfractalsAhlforsconformaldimension

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces a class of self-similar fractals built from finitely many contractive copies of a regular $J$-gon, subject to a local symmetry condition: whenever two cells at any depth share a boundary segment, the two pieces of the fractal inside them are mirror images across that segment. It claims that under several explicit combinatorial conditions on the polygon's boundary segments and the symmetry group, the metric space $(K,d^*)$ is $p$-conductively homogeneous for every $p>\dim_{\mathrm{AR}}(K,d^*)$. Conductive homogeneity is a sufficient condition, introduced in a prior work of the first author, for constructing analogues of $(1,p)$-Sobolev spaces by scaling discrete $p$-energies. For $p=2$ this yields a strongly local regular Dirichlet form, a continuous heat kernel with two-sided sub-Gaussian bounds, and hence a Brownian motion. The family is rich enough to include sets whose global isometry group is trivial, unlike the Sierpinski carpet, unconstrained carpets, and Octa-carpets.

What carries the argument

The argument is carried by conductance constants $E_{M,p,m}(w)$ and neighbour disparity constants $\sigma_{p,m,n}$; $p$-conductive homogeneity means the product $\sup_w E_{M,p,m}(w)\, \sup_n \sigma_{p,m,n}$ stays bounded, which Theorem 5.6 rephrases as geometric two-sided scaling of both quantities, and Theorem 5.9 reduces the condition to the 'knight move' bound $E_{M,p,m}(z)\le c(k) E_{p,m}(u,v,T_k)$. The paper's new work is to verify that bound from geometry: essential boundary segments $(Z_J)_e$, which record which sides of the polygon actually appear as shared sides of cells; a no-isolated-contact-point condition that lets star-paths $E^*_n$ be replaced by edge-paths $E^{\ell}_n$ with bounded modulus loss; and, for low-symmetry cases, a boundary-correspondence map $F_{\partial}$ and a folding map that transfer paths between levels.

What would settle it

Directly test the equivalent knight-move condition on a concrete low-symmetry candidate, for instance the trivial-$G$ pentagon configuration of Example 13.12: compute $E_{M,p,m}(z)$ for a central cell $z$ and $E_{p,m}(u,v,T_k)$ for pairs far apart, and look for an unbounded ratio as $m$ grows for some $p>\dim_{\mathrm{AR}}(K,d^*)$; an unbounded ratio would disprove $p$-conductive homogeneity for that example.

Watch

Extended reading notes

Core claim

For a $G$-symmetric $J$-gon-based self-similar system satisfying one of the paper's sufficient conditions --- $J=3$ with no further assumptions, $G$ acting transitively on the full boundary index set $Z_J$, essential boundary segments $(Z_J)_e$ transitive when isolated contact points are absent, certain even-$J$ cases with $G=D_q$ or $G=\mathrm{Rot}_q$ governed by boundary-correspondence data, or the trivial-$G$ case governed by iterating a finite map $F_{\partial}$ on subsets of boundary segments --- the associated self-similar set $(K,d^*)$ is $p$-conductively homogeneous for every $p>\dim_{\mathrm{AR}}(K,d^*)$. A direct corollary is existence of Banach spaces $W_p$ that mimic $(1,p)$-Sobolev spaces, including density in continuous functions, Markov property, self-similar energy identity, and Hölder-type bounds. In the case $p=2$, $(\widehat{E}_2,W_2)$ is a strongly local regular Dirichlet form and the associated diffusion has a continuous heat kernel satisfying two-sided sub-Gaussian estimates; this gives Brownian motions on infinitely ramified self-similar sets whose global symmetry group can be trivial.

Load-bearing premise

The load-bearing assumption is that cells have no isolated contact points---single vertex intersections that disconnect the union of neighbouring cells---because that is what allows $E^*_n$-paths to be replaced by $E^{\ell}_n$-paths with only a bounded modulus loss, and without it the combinatorial route to the knight-move condition is not established.

Editorial extensions

If this is right

Every $(J,G)$-s.s. system with $J=3$ is $p$-conductively homogeneous for $p>\dim_{\mathrm{AR}}(K,d^*)$, regardless of global symmetry, so the triangle-based family yields Sobolev spaces and diffusions without any symmetry assumptions.
At $p=2$, conductive homogeneity produces a strongly local regular Dirichlet form with continuous heat kernel and two-sided estimates, giving Brownian motions on infinitely ramified polygon-based fractals whose isometry group may be trivial.
The construction yields $W_p$ spaces with Markov property, density in $C(K)$, self-similar energy identity, and Hölder-type bounds, extending Sobolev-theoretic tools to fractals where upper-gradient methods fail.
For even $J$, the boundary-correspondence conditions give a checkable finite criterion: whether certain subsets of boundary indices appear as $F_{\partial}$-images decides conductive homogeneity for $G=D_q$ or $G=\mathrm{Rot}_q$ systems without isolated contacts.
For trivial $G$, the finite iteration of $F_{\partial}$ gives an algorithmic criterion (the three equivalent conditions $F_{\partial}1$--$F_{\partial}3$) that certifies $p$-conductive homogeneity for all $p>\dim_{\mathrm{AR}}(K,d^*)$.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper leaves open whether the no-isolated-contact-point hypothesis in Theorem 6.11 is removable; since Theorems 6.3 and 6.11 are otherwise identical, the technical gap suggests testing transitivity of $(Z_J)_e$ on systems with vertex-only cell intersections.
The $F_{\partial}$-iteration criterion in the trivial-$G$ case is effectively a finite-state automaton on subsets of $Z_J$; one could implement it as a combinatorial pre-screening for conductive homogeneity of a given polygon configuration.
The existence of examples with trivial isometry group suggests conductive homogeneity is a local ratio property rather than a global symmetry property; a natural extension would be non-self-similar locally symmetric tilings obtained by varying contraction ratios cell by cell.
For $p=2$, the two-sided heat kernel estimates contain an exponent $\beta_*$; computing $\beta_*$ numerically for the new examples would connect the construction to the spectral dimension and walk dimension of these diffusions.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

This is a serious paper and the new examples are real. The locally symmetric polygon-based framework, the essential boundary segments, and the F_d iteration criterion in Section 13 are genuinely new relative to the literature, and they give the first family of infinitely ramified self-similar sets with trivial global isometry group for which p-conductive homogeneity is claimed. If the proofs are completed, the Brownian motion consequences are substantial. The paper also does something useful: it reduces conductive homogeneity to finite combinatorial checks on paths and boundary segments, and the Jekyll/Hyde and no-symmetry examples show the criterion has nontrivial content.

The stress-test concern about Theorem 8.8 is correct. In the general case the proof only treats the branch where every connected subset A of the displayed union has #A <= 2, invokes Lemma 8.10, and then says "Applying the above argument". The complementary branch, where #A >= 3, is not analyzed. That matters: a connected set of three or more cells can have as few as two boundary segments, so the desired conclusion does not follow automatically. Since Theorem 8.8 is used in the proofs of Theorems 6.1, 6.3, 6.11, 10.2, and 13.9, the central sufficient conditions are not established as written. This is a load-bearing gap, not a cosmetic one.

There is a second issue: reference [15] is a private communication used for the existence of a folding map when J = 5. A referee cannot check that. The authors should either include the argument or state explicitly which results depend on it. The heavy reliance on the first author's monograph [13] is fine, since that is published prior theory. The example checks done by inspecting figures are less concerning, but given the combinatorial depth an explicit finite verification for the examples would be appropriate.

The right audience is the analysis-on-fractals and metric-measure community, especially people working on diffusions on carpets. My overall sense is that the framework is valuable and the examples are worth taking seriously, but the proof is incomplete. I would send the paper to a serious referee because the gap looks fixable and the contribution would be important. I would not accept it in this form.

Referee Report

2 major / 6 minor

Summary. The paper introduces a class of self-similar sets called G-symmetric J-gon-based self-similar systems and gives sufficient conditions, in terms of the group G and the essential boundary segments (Z_J)_e, for the associated self-similar set (K,d*) to be p-conductively homogeneous for every p > dim_AR(K,d*). Conductive homogeneity is a sufficient condition, developed by the first author in [13], for constructing Sobolev-type spaces W_p and, when p=2, a strongly local regular Dirichlet form with continuous heat kernel and two-sided estimates, hence a Brownian motion. The main theorems are Theorem 6.1 (J=3, no extra condition), Theorem 6.3 (G-transitive action on Z_J), Theorem 6.11 (G-transitive action on (Z_J)_e, no isolated contact points), Theorem 10.2 (even J, G=Rot_q or D_q, a contrapositive condition in terms of boundary components), and Theorem 13.9 (G={I}, no isolated contact points, a condition on iterations of the boundary map F_∂). The proofs reduce conductive homogeneity to the knight-move condition of Theorem 5.9 through two backbone theorems, Theorems 7.2 and 8.8, which assert the existence of connected sets with sufficiently large essential boundary in certain unions of paths.

Significance. If correct, the paper substantially enlarges the class of infinitely ramified self-similar sets known to admit Brownian motions and p-energy Sobolev spaces, and it does so in a direction that is novel: the examples may have trivial global isometry group, in contrast to the Sierpinski carpet, unconstrained carpets, and the Octa-carpet. The sufficient conditions are expressed in finite combinatorial data (the action of G on Z_J and the iterates of F_∂), making them checkable for concrete configurations, as demonstrated in Examples 10.3–10.5 and 13.12–13.13. The paper builds on the published theory of [13] rather than re-deriving it, and the overall proof architecture via conductance constants, neighbor disparity constants, and the knight-move condition is clear and well organized. However, the central combinatorial backbone theorem, Theorem 8.8, contains a proof gap (see Major Comment 1), and a load-bearing existence assertion for J=5 rests on a private communication (see Major Comment 2); these issues must be resolved before the results can be considered established.

major comments (2)

[Section 8, proof of Theorem 8.8] In the proof of Theorem 8.8, the 'general case' after Lemma 8.9 is only carried out under the supposition that every (T_m,E*_m)-connected subset A of the union in (8.2) has #(A)≤2; this is exactly the hypothesis of Lemma 8.10, which then reduces [γ]_n to an E^ℓ_n-path. The complementary case, where some connected A has #(A)≥3, is not treated. The theorem's conclusion requires a specific boundary condition (#(∂A)≥3 or ∂A={i,i+J/2}), and a connected set of three or more cells can have as few as two boundary segments (for example, a linear chain ending on one boundary segment). Consequently, the general case of Theorem 8.8 is not established as written. Since Theorem 8.8 is invoked in the proofs of Theorems 6.1, 6.3, 6.11, 10.2, and 13.9, the central sufficient conditions for conductive homogeneity are not fully supported.
[Section 13, paragraph before Proposition 13.2] The assertion that 'there always exists a folding map for J=5' is attributed to [15], a private communication by K. Sasaya. This assertion is used to claim that one may assume G={I} without loss of generality if J=5. A private communication is not a citable proof for a mathematical existence statement. The authors should either prove this assertion, replace it with a public reference, or explicitly state that the paper's J=5 results do not depend on it; as written, the claim of generality for the J=5 case rests on unverifiable grounds.

minor comments (6)

[Section 8, proof of Theorem 8.8] The expression '#((∂σF_n(bγ))e)⊆{α1,α2}' appears to contain a misplaced #; it should read '(∂σF_n(bγ))e ⊆ {α1,α2}'.
[Section 11, title] The title 'Bondary correspondence' should be 'Boundary correspondence'.
[Section 13, Proposition 13.2(1)] In Proposition 13.2(1), 'holding map' should be 'folding map'.
[Section 6, Theorem 6.11] The statement of Theorem 6.11 says 'no contact point of cells'; this should be 'no isolated contact point of cells' to match Definition 4.1 and the surrounding text.
[Section 2, Example 2.15] In Example 2.15, the phrase 'the corresponding octagons' should read 'the corresponding pentagons', since the example concerns J=5.
[Section 4, proof of Lemma 4.3(2)] In the proof of Lemma 4.3(2), the sentence 'Therefore, (Γ_n(x), E^ℓ_n|Γ_n(x)) is disconnected for any n≥1' should presumably say 'connected for any n≥1' (or 'for some n'), given the logical structure of the proof.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the new combinatorial sufficient conditions are proven against the externally established [13] criterion, with no fitted parameters or assumed conclusion.

full rationale

The derivation chain is not circular. The target property, p-conductive homogeneity, is imported from [13] only as a definition and as an external criterion (Theorem 5.9, the knight-move condition); the paper never assumes that the specific (J,G)-systems are conductively homogeneous. The new work consists of combinatorial and topological sufficient conditions (Theorems 6.1, 6.3, 6.11, 7.2, 8.8, 10.2, 13.9) that construct connected sets A with prescribed essential boundary segments inside H(γ), and then use [13]'s modulus and conductance lemmas to verify the knight-move bound. No parameter is fitted to the target quantity, and no definition is formulated in terms of the conclusion. The reliance on the first author's prior monograph [13] is real but external: its assumptions do not include the conductive homogeneity of the new family, and the cited lemmas are stated independently of the present examples. The reported gap in the proof of Theorem 8.8 is a completeness or correctness concern, not a circular reduction.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No parameters are fitted to data. The theorems quantify over the family of contraction systems, so the contraction ratio and symmetry group are hypotheses rather than fitted constants. The main unproved external inputs are the prior conductive homogeneity theory and the private-communication folding map assertion.

assumptions (4)

standard math Standard iterated function system theory, including existence of the self-similar set and the open set condition dimension formula.
Proposition 2.7 invokes [11, Theorem 1.1.4] and [11, Corollary 1.5.9]; the open set condition follows from (A4), which forces disjoint interiors of first-level cells.
domain assumption The conductive homogeneity framework of [13], including the energy construction, Theorem 5.6, Theorem 5.9, and Lemma C.4.
The paper imports the definition and the main machinery from the first author's prior monograph instead of re-deriving them. This is published theory, not an assumption of the paper's conclusion.
domain assumption No isolated contact point of cells holds for the systems covered by Theorems 6.11, 10.2, and 13.9.
This condition is used to replace E*_n paths by E^ell_n paths in Lemma 7.6 and to control path neighborhoods. For concrete examples it is verified by finite inspection of first-level configurations.
ad hoc to paper For J=5, a folding map always exists, attributed to [15], a private communication by K. Sasaya.
This justifies the remark that one may assume G={I} for J=5. The proof is not given in the paper and cannot be independently checked by the reader.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Conductive homogeneity of locally symmetric polygon-based self-similar sets." pith.science (2026). https://pith.science/paper/NYHANMD2

@misc{pith2026250523057,
  author       = {Pith},
  title        = {Pith review of: Conductive homogeneity of locally symmetric polygon-based self-similar sets},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NYHANMD2}},
  note         = {Machine review of arXiv:2505.23057}
}

read the original abstract

We provide a rich family of self-similar sets, called locally symmetric polygon-based self-similar sets, as examples of metric spaces having conductive homogeneity, which was introduced as a sufficient condition for the construction of counterparts of "Sobolev spaces" on compact metric spaces. In particular, our results imply the existence of "Brownian motions" on our family of self-similar sets at the same time. Unlike the known examples like the Sierpinski carpet by Barlow-Bass, unconstrained carpet by Cao and Qiu and the Octa-carpet by Andrews, our examples may have no global symmetries, i.e. the group of isometries is trivial.

Figures

Figures reproduced from arXiv: 2505.23057 by the authors.

**Figure 1.** Locally symmetric J-gon-based self-similar sets In fact, a variety of analyses on metric spaces, connected with constructions of “Sobolev spaces” in one way or another, has been developed since the late 1980s, partly because the notion of fractal had emerged as models of shapes in nature. In general, fractals do not carry “differentiable” structures, so that conventional analysis based on differentiation can not be … view at source ↗

**Figure 2.** J = 8, G = Rot4 Since our examples are planar, it is easy to see that 2 > dimAR(X, d), so that they are 2-conductive homogeneous under any of sufficient conditions for p-conductive homogeneity for all p > dimAR(X, d). Moreover, 2-conductive homogeneity implies the existence of non-trivial self-similar Dirichlet forms. See Theorem 5.8 for details. Consequently, our results in this paper provide new examples of self-s… view at source ↗

**Figure 3.** Local Symmetry Octagon for the Octa-carpet. Our Brownian motions constructed through 2- conductive homogeneity provide new examples of those on regular polygonbased infinitely-ramified self-similar sets which do not necessarily inherit the full symmetry from the regular polygon. For simplicity, we are going to explain exact definitions and results in the case of hexagon-based self-similar sets. Let Q∗ be a regular … view at source ↗

Figures from the paper (16 more)

**Figure 4.** Figure 4: Hexagonal cases In the case when G = D3, the essential boundary segments (Z6) e may not be Z6 any longer. In fact, G = D3 and (Z) e = {1, 3, 5} for the example in [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

**Figure 5.** Figure 5: Q (5) ∗ and Q (6) ∗ Definition 2.2. Define Rθ as the reflection in the line {(t cos θ, tsin θ)|t ∈ R} and define Θθ as the rotation about the origin 0 by an angle θ. Definition 2.3. Define DJ = {g|g ∈ O(2), g(Q (J) ∗ ) = Q (J) ∗ }. and RotJ = {Θ2π J i |i ∈ ZJ }. It is …

**Figure 6.** Figure 6: J = 3, G = {I} for any s ∈ S. Then (S, {fs}s∈S, G) is a (8, G)-s.s. systems with G = D8. The associated self-similar set K is called the Octa-gasket in [2] or Octa-carpet in [3]. We will revisit this example as Example 6.7, where the p-conductively homogeneity of K is …

**Figure 7.** Figure 7: Octa-carpet: J = 8, G = D4 Example 2.15. Let J = 5. Define S as the collection of small white and grey pentagons in [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]

**Figure 1.** Figure 1: Pentagon-base: Figure 8: J = 5G [PITH_FULL_IMAGE:figures/full_fig_p015_1.png]

**Figure 9.** Figure 9: J = 8, G = {DV 4 } The next lemma tells us basic properties of an isolated contact point of cells. Lemma 4.3. Let (S, {fs}s∈S, G) be a (J, G)-self-similar system with J ≥ 3 and let K be the associated self-similar set. (1) If x ∈ K and Qn(x)\{x} is disconnected, then, …

**Figure 10.** Figure 10 [PITH_FULL_IMAGE:figures/full_fig_p031_10.png]

**Figure 11.** Figure 11: (a). As before, the contraction ratio is determined by the configuration of s ∈ S, and cs is the centre of the octagon s. The map φs is defined as I, Θπ, Θ− 2π 9 , R2π 9 , Θ2π 9 and R− 2π 9 respectively if the corresponding nonagon s in [PITH_FULL_IMAGE:figures/full…

**Figure 12.** Figure 12: J = 6 and G = DV 3 Theorem 6.11. Assume that there exists no contact point of cells and that (ZJ ) e is G-transitive. Then (K, d∗) is p-conductively homogeneous for any p > dimAR(K, d∗). Note that if ZJ is G-transitive as we assumed in Theorem 6.3, then (ZJ ) e = ZJ a…

**Figure 13.** Figure 13: J = 6, G = D3 (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p057_13.png]

**Figure 14.** Figure 14: ConT (Z 0 6 ) and ConT (Z 1 6 ) ConT (Z 0 6 ) and ConT (Z 1 6 ) for these two isomers in [PITH_FULL_IMAGE:figures/full_fig_p057_14.png]

**Figure 15.** Figure 15: Hexagon-based isomers: J = 6, G = Rot3 b1 b0 b2 b3 b4 b5 b0 b1 b2 b3 b4 b5 [PITH_FULL_IMAGE:figures/full_fig_p058_15.png]

**Figure 16.** Figure 16: ConT (Z 0 J ) and ConT (Z 1 J ) Now Theorem 10.2 shows that (K, d∗) is p-conductively homogeneous for any p > dimAR(K, d∗). Case Hyde( [PITH_FULL_IMAGE:figures/full_fig_p058_16.png]

**Figure 17.** Figure 17: J = 5, G = {I} Example 13.12. Let S be the collection of pentagons in the most left figure of [PITH_FULL_IMAGE:figures/full_fig_p072_17.png]

**Figure 18.** Figure 18: Con [PITH_FULL_IMAGE:figures/full_fig_p073_18.png]

**Figure 19.** Figure 19: J = 6, G = {I} Example 13.13. Let S be the collection of hexagons in the most left figure of [PITH_FULL_IMAGE:figures/full_fig_p073_19.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

16 extracted references · 15 canonical work pages

[15]

Sasaya, private communication

K. Sasaya, private communication

work page
[13]

European Math

,Conductive homogeneity of compact metric spaces and construction ofp-energy, Mem. European Math. Soc.5(2023)

work page 2023
[1]

M. T. Barlow and R. F. Bass,The construction of Brownian motion on the Sierpinski carpet, Ann. Inst. Henri Poincar´ e25(1989), 225–257

work page 1989
[2]

Berry, S

T. Berry, S. Hellman, and R. Strichartz,Outer approximation of the spec- trum of a fractal Laplacian, Experimental Math.18(2009), 449–480

work page 2009
[3]

Canner, C

C. Canner, C. Hayes, R. Huang, M. Orwin, and L. G. Rogers,Resistance scaling on 4N-carpets, Forum Mathematicum34(2022), 61–75

work page 2022
[4]

Cao and H

S. Cao and H. Qiu,Dirichlet forms on unconstrained Sierpinski carpets, Probab. Theory Relat. Fields189(2024), 613–657

work page 2024
[5]

Cheeger,Differentiability of Lipschitz functions on metric measure spaces, Geom

J. Cheeger,Differentiability of Lipschitz functions on metric measure spaces, Geom. Funct. Anal.3(1999), 428–517

work page 1999
[6]

Haj lasz,Sobolev spaces on an arbitrary metric spaces, Pontential Anal- ysis5(1996), 403–415

P. Haj lasz,Sobolev spaces on an arbitrary metric spaces, Pontential Anal- ysis5(1996), 403–415. 82

work page 1996

Show all 16 references

[7]

Heinonen, P

J. Heinonen, P. Koskela, N. Shanmugalingam, and J. Tyson,Sobolev Spaces on Metric Measure Spaces - An Approach Based on Upper Gradients, New Mathematical Monographs, Cambridge University Press, 2015

2015
[8]

U. A. Andrews IV,Existence of diffusions on 4N-carpets, Doctoral Disser- tation, University of Connecticut, 2017

2017
[9]

Kajino and M

N. Kajino and M. Murugan,On the conformal walk dimension II: Non- attainment for some Sierpi´ nski carpets, in preparation
[10]

Math.231(2023), 263–405

,On the conformal walk dimension: Quasisymmetric uniformization for symmetric diffusions, Invent. Math.231(2023), 263–405

2023
[11]

Kigami,Analysis on Fractals, Cambridge Tracts in Math

J. Kigami,Analysis on Fractals, Cambridge Tracts in Math. vol. 143, Cam- bridge University Press, 2001

2001
[12]

,Geometry and Analysis of Metric Spaces via Weighted Partitions, Lecture Notes in Math. vol. 2265, Springer, 2020

2020
[14]

Ota,Conductive homogeneity of self-similar sets; a new class of exam- ples, Doctoral Dissertation, Kyoto University, 2025

Y. Ota,Conductive homogeneity of self-similar sets; a new class of exam- ples, Doctoral Dissertation, Kyoto University, 2025

2025
[16]

Shanmugalingam,Newtonian spaces: an extension of Sobolev spaces to metric measure spaces, Rev

N. Shanmugalingam,Newtonian spaces: an extension of Sobolev spaces to metric measure spaces, Rev. Mat. Iberoamer.16(2000), 243–279. 83

2000

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [15]

Sasaya, private communication

K. Sasaya, private communication

work page

[2] [13]

European Math

,Conductive homogeneity of compact metric spaces and construction ofp-energy, Mem. European Math. Soc.5(2023)

work page 2023

[3] [1]

M. T. Barlow and R. F. Bass,The construction of Brownian motion on the Sierpinski carpet, Ann. Inst. Henri Poincar´ e25(1989), 225–257

work page 1989

[4] [2]

Berry, S

T. Berry, S. Hellman, and R. Strichartz,Outer approximation of the spec- trum of a fractal Laplacian, Experimental Math.18(2009), 449–480

work page 2009

[5] [3]

Canner, C

C. Canner, C. Hayes, R. Huang, M. Orwin, and L. G. Rogers,Resistance scaling on 4N-carpets, Forum Mathematicum34(2022), 61–75

work page 2022

[6] [4]

Cao and H

S. Cao and H. Qiu,Dirichlet forms on unconstrained Sierpinski carpets, Probab. Theory Relat. Fields189(2024), 613–657

work page 2024

[7] [5]

Cheeger,Differentiability of Lipschitz functions on metric measure spaces, Geom

J. Cheeger,Differentiability of Lipschitz functions on metric measure spaces, Geom. Funct. Anal.3(1999), 428–517

work page 1999

[8] [6]

Haj lasz,Sobolev spaces on an arbitrary metric spaces, Pontential Anal- ysis5(1996), 403–415

P. Haj lasz,Sobolev spaces on an arbitrary metric spaces, Pontential Anal- ysis5(1996), 403–415. 82

work page 1996