Dyadic norm Besov-type spaces as trace spaces on regular trees

Pekka Koskela; Zhuang Wang

Dyadic norm Besov-type spaces as trace spaces on regular trees

T0 review · 0 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read On any regular tree, the boundary traces of weighted Sobolev functions form exactly a Besov-type space, with bounded linear trace and extension operators.

desk verdict Main trace results are new and mostly solid, but Theorem 1.2's right-inverse claim has a domain mismatch that the stress-test note got backwards. read the letter →

arxiv 1908.06937 v1 pith:IDAA6UAU submitted 2019-08-19 math.FA

Pekka Koskela , Zhuang Wang This is my paper · ORCID

classification math.FA MSC 46E3530L05

keywords Besov-typespaceregulartreetracedyadicnormNewtonianweightedSobolevAhlforsmeasureCantor-typeboundary

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 proves that on any regular (K-ary) tree, the boundary traces of weighted first-order Sobolev functions are exactly the functions in a Besov-type space defined by a dyadic energy on the tree's Cantor-type boundary. The trace space is $B^{\theta,\lambda}_p(\partial X)$ with $\theta = 1 - (\beta-\log K)/(\varepsilon p)$, where $K$ is the branching number, $\beta$ controls the weight $e^{-\beta|x|}$ on the tree, and $\varepsilon$ determines the visual metric. The identification is quantitative: there is a bounded linear trace operator and a bounded linear extension operator that is its right inverse. The paper also settles the borderline case $p=(\beta-\log K)/\varepsilon$, where the trace picture splits, including a sharp range for $\lambda$ and a different Besov-type space when $p=1$. This provides an exact, computable description of boundary values for weighted Sobolev spaces on trees, analogous to classical Euclidean trace theorems.

What carries the argument

The central object is the dyadic Besov-type energy on the boundary, $$\|f\|^p_{\dot $B^{{\theta,\lambda}}$_p(\partial X)}= \sum_{n=1}^\infty $e^{{\varepsilon n \theta p}}$ n^\$\lambda$ \sum_{I\in Q_n} \nu(I) |f_I - f_{\hat I}|^p,$$ where $Q_n$ lists the boundary sets $I_x$ corresponding to vertices $x$ at distance $n$ from the root, $\hat I$ is the parent of $I$, and $f_I$ is the $\nu$-average of $f$ on $I$. The workhorse identity is the comparability $\int_X |g_{\tilde u}|^p\,d\mu_\lambda \approx \|u\|^p_{\dot B^{\theta,\lambda}_p(\partial X)}$ for the piecewise-affine extension $\tilde u$; it follows from the measure estimates $\mu_\lambda([x,y])\approx e^{-\beta n}n^\lambda$ and $\nu(I)\approx e^{-\varepsilon n Q}=e^{-n\log K}$. This comparability, together with the Ahlfors $Q$-regularity of $\nu$ with $Q=\log K/\varepsilon$, is what makes the trace and extension estimates close.

What would settle it

Take a fixed parameter triple in the theorem's range and a boundary function constant on one level-$n$ dyadic ball $I$ and zero elsewhere. Use the paper's affine extension (3.6)–(3.8) and compute the ratio $$R_n=\frac{\int_X |g_{\tilde u}|^p\,d\mu_\$\lambda$}{$e^{{\varepsilon n\theta p}}$n^\$\lambda$\nu(I)|u_I-u_{\hat I}|^p}.$$ Theorem 1.1 predicts $\sup_n R_n<\infty$ for all choices of $I$; if any allowed parameter choice gives $R_n\to\infty$ along a sequence of dyadic balls, the claimed bounded linear extension fails.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.1. For a $K$-ary tree $X$ with $K\ge 2$, fixed $\beta>\log K$, $\varepsilon>0$, $\lambda\in\mathbb{R}$, and $p\ge 1$ with $p>(\beta-\log K)/\varepsilon$, the Besov-type space $B^{\theta,\lambda}_p(\partial X)$ is the trace space of the weighted Newtonian space $N^{1,p}(X,\mu_\lambda)$ with $\theta = 1-(\beta-\log K)/(\varepsilon p)$. The trace operator sends a Sobolev function to its limits along geodesic rays, and the extension operator interpolates the dyadic averages of a boundary function affinely along edges; both are bounded linear and compose to the identity. In the borderline case $p=(\beta-\log K)/\varepsilon\ge 1$, a bounded linear trace into $L^p(\partial X)$ exists for $\lambda>p-1$ (or $\lambda\ge 0$ when $p=1$), and this range is sharp, while a bounded nonlinear extension from $L^p(\partial X)$ to $N^{1,p}(X)$ exists. For $p=1$ and $\lambda>0$, the trace space of $N^{1,1}(X,\mu_\lambda)$ is the space $B^{0,\lambda}_\alpha(\partial X)$, which strictly contains $B^{0,\lambda}_1(\partial X)$. The fixed Whitney-type extension operator is bounded and linear exactly on $B^{0,\lambda}_p(\partial X)$, and this space is optimal for that operator.

Load-bearing premise

The load-bearing premise is that the boundary measure $\nu$ is Ahlfors $Q$-regular with $Q=\log K/\varepsilon$, meaning every boundary ball $B(\xi,r)$ has measure comparable to $r^Q$; every dyadic estimate in the proof, in particular the comparability between the Sobolev gradient energy and the dyadic Besov energy, uses $\nu(I)\approx e^{-\varepsilon n Q}$.

Editorial extensions

If this is right

At $\lambda=0$ and $0<\theta<1$, the dyadic norm is equivalent to the double-integral Besov norm $B^\theta_{p,p}(\partial X)$, so Theorem 1.1 recovers the known trace description for $N^{1,p}(X)$ without weights.
The explicit linear extension operator gives a constructive way to extend boundary data with finite dyadic energy to the whole tree, with the Newtonian energy controlled by the boundary norm.
In the borderline case $p=(\beta-\log K)/\varepsilon$, the trace operator is bounded into $L^p$ for $\lambda>p-1$ (or $\lambda\ge 0$ if $p=1$), and the range is sharp: for $\lambda=p-1-\delta$, traces can be infinite almost everywhere.
For $p=1$ and $\lambda>0$, the true trace space of $N^{1,1}(X,\mu_\lambda)$ is $B^{0,\lambda}_\alpha$, which is strictly larger than $B^{0,\lambda}_1$; setting $\theta=0$ in the main theorem does not give the correct borderline trace space.
For the fixed Whitney-type extension operator, $B^{0,\lambda}_p$ is the optimal domain: any Banach space of boundary functions on which the operator is bounded and linear embeds into $B^{0,\lambda}_p$.

Reading between the lines

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

The same dyadic trace criterion could be tested on non-regular trees by replacing $K^n$ with the growth function $v(n)$ of the tree; the expected critical exponent would involve $Q=\lim_{n\to\infty}\frac{\log v(n)}{\varepsilon n}$.
The logarithmic counterexample at the critical integrability suggests that similar sharp thresholds for traces should appear in other hyperbolic or negatively-curved settings where radial functions diverge logarithmically.
The splitting between $B^{0,\lambda}_1$ and $B^{0,\lambda}_\alpha$ at $p=1$ indicates that when the trace operator is not surjective, the choice of extension operator genuinely selects the trace space; users needing a linear right inverse should use $B^{0,\lambda}_p$, while the full trace space is larger.
One could verify the dyadic energy characterization computationally on small finite truncated trees, checking whether the ratio between the Newtonian extension energy and the boundary dyadic energy stays bounded as the truncation level grows.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

I read the Koskela–Wang paper on dyadic Besov-type trace spaces on regular trees. The headline: the main trace characterizations look right and are genuinely new, but there is a hole in the second part of Theorem 1.2 that the stress-test note hand-waved past.

What is new: the spaces B^{θ,λ}_p and B^{0,λ}_α, the weighted trace characterizations for Newtonian spaces (Theorems 1.1, 1.3, 1.4), the sharp λ > p−1 condition in Theorem 1.2, and the strict inclusion B_1^{0,λ} ⊊ B_α^{0,λ} with a clean probabilistic example. The dyadic norm approach is elegant, and the proofs of the main trace and extension estimates, especially the dyadic energy comparisons (3.10) and the borderline p = (β−log K)/ε case, are detailed and check out. The paper does a good job of honestly stating what it does not prove, e.g., Remark 3.6 on surjectivity for λ > p−1.

Soft spots. First, the composition T∘E in Theorem 1.2 is not legitimate as written. The trace operator T is defined and bounded on the weighted space N^{1,p}(X, μ_λ). The extension E is constructed in Proposition 3.5 only into the unweighted space N^{1,p}(X). For λ > 0, the weighted space is a proper subset of the unweighted one, not a superset. The stress-test note claims an embedding in the wrong direction. The proof of Proposition 3.5 uses only unweighted estimates (Lemmas 3.3 and 3.4), so it does not show that E(f) lies in the domain of T. For p = 1, λ > 0, Theorem 1.3 tells us T is not surjective onto L^1, so a right inverse from L^1 into the weighted space cannot exist. The statement needs repair: either prove E maps into the weighted space under the stated λ conditions, or restrict the right-inverse claim to B_p^{0,λ}, as Theorem 1.4 already does. This is not fatal to the main theorems, but it is a real gap, not just a clarification. Second, Proposition 2.13 is stated without proof; it only connects the dyadic Besov spaces to the classical ones and is not used elsewhere, so this is minor.

The Ahlfors Q-regularity of the boundary measure is a standard input and I do not see a problem there.

Who this is for: researchers working on trace theorems on metric measure spaces, especially on trees and Cantor-type boundaries. The main results deserve a serious referee. The issue in Theorem 1.2 is localized and fixable, so I would send it to review with a request for major revision.

Referee Report

0 major / 5 minor

Summary. The paper introduces Besov-type spaces B^{θ,λ}_p(∂X) on the boundary of a K-ary regular tree, defined through dyadic energies with weights e^{εnθp}n^λν(I), and proves that for β>logK, ε>0, λ∈R, p≥1 with p>(β−logK)/ε and θ=1−(β−logK)/(εp), B^{θ,λ}_p(∂X) is the trace space of the weighted Newtonian space N^{1,p}(X,μλ), where dμλ=e^{-β|x|}(|x|+C)^λd|x|. The trace and extension operators are explicit and bounded linear (Theorem 1.1). The paper also treats the critical case p=(β−logK)/ε (Theorems 1.2–1.4), including a sharp condition λ>p−1 for traces in L^p(∂X), a nonlinear extension operator from L^p(∂X) to N^{1,p}(X), the identification of the trace of N^{1,1}(X,μλ) with the space B^{0,λ}_α, and the optimality of B^{0,λ}_p(∂X) for a Whitney-type extension operator. The proofs use the Ahlfors Q-regularity of the boundary, dyadic energy estimates, Fubini-type counting on the tree, and a Gagliardo-type gluing construction.

Significance. If Theorem 1.1 holds, it gives an exact dyadic characterization of boundary traces for a family of weighted Sobolev spaces on regular trees, extending the unweighted trace theorem of Björn–Björn–Gill–Shanmugalingam and complementing hyperbolic-filling trace results. The paper's main estimates are explicitly verified: the measure comparisons (3.3), the dyadic equivalence (3.10), the counting identity (3.16), and the norm equivalence (3.25) all check out, and the trace and extension operators are constructive. The main external input is the Ahlfors Q-regularity of the visual boundary measure (Proposition 2.10), which is a standard property of regular trees rather than a fragile assumption. I find no load-bearing error in the proof of Theorem 1.1; the issues below concern statement precision and presentation and are local to the statements of Theorems 1.2–1.3.

minor comments (5)

[Section 1, Theorem 1.2 (second paragraph)] As stated, the identity T∘E=Id is not well-defined because T is introduced as a bounded linear operator from N^{1,p}(X,μλ) to L^p(∂X), while E is constructed into the unweighted space N^{1,p}(X). For λ>0 the weighted space is strictly smaller than N^{1,p}(X), and for λ=0, p>1 the unweighted trace is not bounded on all of N^{1,p}(X); indeed Example 3.2 specialized to λ=0 gives an N^{1,p}(X) function whose geodesic limits are infinite. The proof of Proposition 3.5 actually establishes that the limit of E f along rays equals f for ν-a.e. ξ. Please restate the theorem accordingly, e.g., by declaring the trace operator on the union of the relevant spaces or by phrasing the right-inverse property as a pointwise trace identity rather than as T∘E on the full unweighted space. Also, the notation 'Lp(X)' at the end of the statement should be 'Lp(∂X)'.
[Section 2, Proposition 2.13] This proposition is stated without proof. It is used in the introduction to justify the claim that Theorem 1.1 recovers the classical Besov trace results from [3] when λ=0, while not being used in the proofs of Theorems 1.1–1.4. Please either supply the proof, give a precise reference for the equivalence, or explicitly downgrade the recovery claim to a remark so that the omission does not leave a labeled proposition unsupported.
[Section 3.3, proof of Theorem 1.3] In the displayed chain estimating the B^{0,λ}_α energy, the factors α(n) and α(n+1) are missing the exponent λ in two of the sums; the subsequent line uses α(n+1)^λ and shows the intended expression. This should be corrected to α(n)^λ and α(n+1)^λ for consistency.
[Section 3.2, proof of Proposition 3.1] The phrase 'for any λ > p−1 if p = 1 or for λ ≥ 0 if p = 1' is garbled; it should read 'for λ > p−1 if p > 1, or λ ≥ 0 if p = 1'.
[Throughout] There are numerous small typos and OCR artifacts (e.g., 'A /greaterorsimilarB' in Section 1, 'classﬁcation' in the MSC line, 'Haj/suppress lasz' in the references). A careful proofreading pass is needed, but these do not affect the mathematics.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the trace and extension estimates are proven from the definitions with independent Ahlfors-regularity input from [3].

full rationale

I walked the proofs of Theorems 1.1–1.4 and found no load-bearing step that reduces to its own input. The Besov-type space B^{θ,λ}_p(∂X) is defined via its own dyadic energy (Definition 2.12), and the Newtonian space N^{1,p}(X,μ_λ) is defined independently via upper gradients (Section 2.3). The theorem does not assume the trace equality; it proves both inequalities. The trace estimate is obtained from the upper-gradient integral inequality, Fubini's theorem, Hölder's inequality, and the Ahlfors Q-regularity of the boundary measure from [3, Lemma 5.2]. The extension estimate is obtained by constructing a piecewise affine extension from dyadic averages and comparing the gradient energy to the dyadic energy in equation (3.10); the comparison is an algebraic consequence of ν(I) ≈ e^{-εnQ}, Q = logK/ε, and the stated relation θ = 1 − (β−logK)/(εp). This is parameter matching, not circularity: the theorem would fail if the trace estimates did not close, and the proof supplies the missing estimates. The self-citations [20] and [23] are used for motivation and for dyadic/Whitney techniques, not as substitutes for the main arguments; [23] is cited in Proposition 2.13, but that proposition is explicitly marked with 'We omit the details' and is not used in the proofs of Theorems 1.1, 1.2, 1.3, or 1.4. The manuscript itself flags its limitations honestly: Remark 3.6 records the open surjectivity question, Theorem 1.4's optimality is explicitly restricted to the particular operator E, and Proposition 2.13's omitted proof is not load-bearing. There is also no renamed known result: the Besov-type spaces are new objects, and their relation to the classical Besov spaces in Proposition 2.13 is a comparison, not a relabeling. Ahlfors regularity of ∂X is imported from the independent work [3] and is used as an external benchmark, which is legitimate support rather than circular evidence. Overall, the derivation chain is self-contained once the standard boundary-measure property is accepted, and no circular step is exhibited.

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

The central theorems are proven from the definitions with explicit constants. The paper does not fit any parameters to data. It relies on standard Newtonian-space theory, the Ahlfors regularity of ∂X (quoted from [3]), the ball/half-ball comparison lemmas from [3], and one quoted but unproved equivalence (Proposition 2.13). No new physical entities are postulated.

assumptions (4)

domain assumption Boundary ∂X with visual metric is Ahlfors Q-regular with Q = logK/ε
Quoted as Proposition 2.10 from [3, Lemma 5.2]. Used throughout to estimate ν(I) ≈ e^{−ε n Q} and in Proposition 3.1, Theorem 1.3 trace estimates. If the boundary were not Ahlfors regular, the dyadic normalization would break.
domain assumption Ball/half-ball measure comparison estimates (Lemma 2.2 from [3]) and the resulting ball mass asymptotics
Lemma 2.2 is quoted from [3, Lemma 3.1, 3.2]; it is the basis for Corollary 2.4, 2.6, 2.8 and the doubling of μ_λ. These are needed for every norm estimate in Section 3.
standard math Newtonian N^{1,p} space theory: upper gradients, Banach structure, trace conventions
Standard background from Heinonen-Koskela-Shanmugalingam-Tyson [17] and Björn-Björn [2], cited in Section 2.3. The paper relies on the equivalence of the geodesic upper gradient definition with the usual one.
domain assumption Dyadic Besov-type space B^{θ,0}_p coincides with the double-integral Besov space B^θ_{p,p} (Proposition 2.13)
Stated with proof omitted ('We omit the details'). Used to recover the unweighted result from [3] for λ=0. A proof is needed for full self-containedness.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Dyadic norm Besov-type spaces as trace spaces on regular trees." pith.science (2026). https://pith.science/paper/IDAA6UAU

@misc{pith2026190806937,
  author       = {Pith},
  title        = {Pith review of: Dyadic norm Besov-type spaces as trace spaces on regular trees},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IDAA6UAU}},
  note         = {Machine review of arXiv:1908.06937}
}

read the original abstract

In this paper, we study function spaces defined via dyadic energies on the boundaries of regular trees. We show that correct choices of dyadic energies result in Besov-type spaces that are trace spaces of (weighted) first order Sobolev spaces.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 29 canonical work pages

[1]

Aronszajn: Boundary values of functions with ﬁnite Dirichlet integral , Techn

N. Aronszajn: Boundary values of functions with ﬁnite Dirichlet integral , Techn. Report 14, University of Kansas, 1955

work page 1955
[2]

Bj ¨orn and J

A. Bj ¨orn and J. Bj ¨orn: Nonlinear potential theory on metric spaces , EMS Tracts Math. 17, European Mathematical Society, Zurich 2011

work page 2011
[3]

Bj ¨orn, J

A. Bj ¨orn, J. Bj¨orn, J. T. Gill and N. Shanmugalingam: Geometric analysis on Cantor sets and trees. J. Reine Angew. Math. 725 (2017), 63-114

work page 2017
[4]

Bj ¨orn, J

A. Bj ¨orn, J. Bj ¨orn and N. Shanmugalingam: The Dirichlet problem for p- harmonic functions on metric spaces , J. Reine Angew. Math. 556 (2003), 173-203

work page 2003
[5]

M. Bonk, J. Heinonen and P. Koskela: Uniformizing Gromov hyperbolic spaces, Ast ˜Al’risque No. 270 (2001), viii+99 pp

work page 2001
[6]

Bonk and E

M. Bonk and E. Saksman: Sobolev spaces and hyperbolic ﬁllings , J. Reine Angew. Math. 737 (2018), 161-187

work page 2018
[7]

Bridson and A

M. Bridson and A. Haeﬂiger: Metric spaces of non-positive curvature, Grundlehren Math. Wiss. 319, Springer-Verlag, Berlin 1999

work page 1999
[8]

V. I. Burenkov and M. L. Goldman: Extension of functions from Lp, Studies in the theory of diﬀerentiable functions of several variables and its applica- tions, VII, Trudy Mat. Inst. Steklov. 150 (1979), 31-51, 321

work page 1979

Show all 30 references

[9]

Farkas, J

W. Farkas, J. Johnsen and W. Sickel: Traces of anisotropic Besov-Lizorkin- Triebel spaces-a complete treatment of the borderline case s, Math. Bohem. 125 (2000), no. 1, 1-37

2000
[10]

Gagliardo: Caratterizzazioni delle tracce sulla frontiera relative a d alcune classi di funzioni in n variabili, Rend

E. Gagliardo: Caratterizzazioni delle tracce sulla frontiera relative a d alcune classi di funzioni in n variabili, Rend. Sem. Mat. Univ. Padova 27 (1957), 284–305

1957
[11]

Ginzburg, Traces of functions from weighted classes , Izv

A. Ginzburg, Traces of functions from weighted classes , Izv. Vyssh. Uchebn. Zaved. Mat. (1984) 61ˆ a˘A¸ S64

1984
[12]

Haj/suppress lasz:Sobolev space on metric-measure spaces, in Heat kernels and analysis on manifolds, graphs and metric spaces (Paris 2002 ), Contemp

P. Haj/suppress lasz:Sobolev space on metric-measure spaces, in Heat kernels and analysis on manifolds, graphs and metric spaces (Paris 2002 ), Contemp. Math. 338, American Mathematical Society, Providence (2003), 1 73-218

2003
[13]

Haj/suppress lasz and P

P. Haj/suppress lasz and P. Koskela:Sobolev met Poincar´ e, Mem. Amer. Math. Soc. (2000), no. 688, x+101 pp. 36 P. Koskela, Z. Wang

2000
[14]

Haroske and H

D. Haroske and H. J. Schmeisser: On trace spaces of function spaces with a radial weight: the atomic approach , Complex Var. Elliptic Equ. 55 (2010), no. 8-10, 875-896

2010
[15]

Heinonen: Lectures on analysis on metric spaces , Universitext, Springer- Verlag, New York 2001

J. Heinonen: Lectures on analysis on metric spaces , Universitext, Springer- Verlag, New York 2001

2001
[16]

Heinonen and P

J. Heinonen and P. Koskela: Quasiconformal mappings in metric spaces with controlled geometry, Acta Math. 181 (1998), 1-61

1998
[17]

Heinonen, P

J. Heinonen, P. Koskela, N. Shanmugalingam and J. Tyson: Sobolev Spaces on Metric Measure Spaces: An Approach Based on Upper Gradien ts. Cam- bridge: Cambridge University Press, 2015

2015
[18]

Johnsen: Traces of Besov spaces revisited , Z

J. Johnsen: Traces of Besov spaces revisited , Z. Anal. Anwendungen 19 (2000), no. 3, 763-779

2000
[19]

Jonsson and H

A. Jonsson and H. Wallin: The trace to subsets of Rn of Besov spaces in the general case, Anal. Math. 6 (1980), 223-254

1980
[20]

Kauranen, P

A. Kauranen, P. Koskela and A. Zapadinskaya: Regularity and Modulus of Continuity of Space-Filling Curves, to appear in J. Analyse Math
[21]

Mal´ y:Trace and extension theorems for Sobolev-type functions in metric spaces, arXiv:1704.06344

L. Mal´ y:Trace and extension theorems for Sobolev-type functions in metric spaces, arXiv:1704.06344

arXiv
[22]

Mal´ y, N

L. Mal´ y, N. Shanmugalingam and M. Snipes:Trace and extension theorems for functions of bounded variation , Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) 18 (2018), no. 1, 313-341

2018
[23]

Koskela, T

P. Koskela, T. Soto and Z. Wang: Traces of weighted function spaces: dyadic norms and Whitney extensions. Sci. China Math. 60 (2017), no. 11, 1981- 2010

2017
[24]

Peetre: A counterexample connected with Gagliardo’s trace theorem, Spe- cial issue dedicated to W˚ A´Cadys˚A´Caw Orlicz on the occasion of his seventy- ﬁfth birthday, Comment

J. Peetre: A counterexample connected with Gagliardo’s trace theorem, Spe- cial issue dedicated to W˚ A´Cadys˚A´Caw Orlicz on the occasion of his seventy- ﬁfth birthday, Comment. Math. Special Issue 2 (1979), 277-282

1979
[25]

Saksman and T

E. Saksman and T. Soto: Traces of Besov, Triebel-Lizorkin and Sobolev spaces on metric spaces , Anal. Geom. Metr. Spaces 5 (2017), 98-115

2017
[26]

Soto: Besov spaces on metric spaces via hyperbolic ﬁllings , arXiv:1606.08082

T. Soto: Besov spaces on metric spaces via hyperbolic ﬁllings , arXiv:1606.08082

arXiv
[27]

Triebel: Theory of function spaces , Monographs in Mathematics, 78

H. Triebel: Theory of function spaces , Monographs in Mathematics, 78. Birkh¨auser Verlag, Basel, 1983. Dyadic norm Besov-type spaces as trace spaces on regular trees 37

1983
[28]

Triebel: The structure of functions , Monographs in Mathematics, 97

H. Triebel: The structure of functions , Monographs in Mathematics, 97. Birkh¨auser Verlag, Basel, 2001

2001
[29]

A. I. Tyulenev: Description of traces of functions in the Sobolev space with a Muckenhoupt weight , Proc. Steklov Inst. Math. 284 (2014), no. 1, 280–295

2014
[30]

A. I. Tyulenev: Traces of weighted Sobolev spaces with Muckenhoupt weight. The case p = 1, Nonlinear Anal. 128 (2015), 248–272. Pekka Koskela Department of Mathematics and Statistics, University of Jyv ¨askyl¨a, PO Box 35, FI-40014 Jyv¨askyl¨a, Finland E-mail address: pekka.j....

2015

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Aronszajn: Boundary values of functions with ﬁnite Dirichlet integral , Techn

N. Aronszajn: Boundary values of functions with ﬁnite Dirichlet integral , Techn. Report 14, University of Kansas, 1955

work page 1955

[2] [2]

Bj ¨orn and J

A. Bj ¨orn and J. Bj ¨orn: Nonlinear potential theory on metric spaces , EMS Tracts Math. 17, European Mathematical Society, Zurich 2011

work page 2011

[3] [3]

Bj ¨orn, J

A. Bj ¨orn, J. Bj¨orn, J. T. Gill and N. Shanmugalingam: Geometric analysis on Cantor sets and trees. J. Reine Angew. Math. 725 (2017), 63-114

work page 2017

[4] [4]

Bj ¨orn, J

A. Bj ¨orn, J. Bj ¨orn and N. Shanmugalingam: The Dirichlet problem for p- harmonic functions on metric spaces , J. Reine Angew. Math. 556 (2003), 173-203

work page 2003

[5] [5]

M. Bonk, J. Heinonen and P. Koskela: Uniformizing Gromov hyperbolic spaces, Ast ˜Al’risque No. 270 (2001), viii+99 pp

work page 2001

[6] [6]

Bonk and E

M. Bonk and E. Saksman: Sobolev spaces and hyperbolic ﬁllings , J. Reine Angew. Math. 737 (2018), 161-187

work page 2018

[7] [7]

Bridson and A

M. Bridson and A. Haeﬂiger: Metric spaces of non-positive curvature, Grundlehren Math. Wiss. 319, Springer-Verlag, Berlin 1999

work page 1999

[8] [8]

V. I. Burenkov and M. L. Goldman: Extension of functions from Lp, Studies in the theory of diﬀerentiable functions of several variables and its applica- tions, VII, Trudy Mat. Inst. Steklov. 150 (1979), 31-51, 321

work page 1979