REVIEW 2 major objections 4 minor 1 cited by
Weighted minimum $\alpha$-Green energy problems
T0 review · 2 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash
Pith's one-line read For α-Green kernels, the weighted minimum-energy problem has a minimizer exactly when the external-field measure has total mass at least 1, and at mass 1 the minimizer is the α-Green balayage of that measure onto the constraint set.
desk verdict Solid extension of the author's own program to α-Green kernels, with a genuinely useful existence criterion, but the 'if and only if' in Theorem 2.4 rests on one imported theorem that is not re-proved here. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the $\alpha$-Green kernel $g^\alpha_D(x,y):=\kappa_\alpha(x,y)-U_{\kappa_\alpha}^{(\varepsilon_x)^Y_{\kappa_\alpha}}(y)$, the $\alpha$-Riesz kernel minus the potential of the balayage of the unit Dirac mass onto the complement of $D$. For $\alpha\leqslant2$ this kernel is perfect, meaning the cone of positive finite-energy measures is strongly complete and strong convergence implies vague convergence, so minimizing sequences have strong limits. The argument also uses the $\alpha$-Green balayage $\vartheta^F_g$, the measure on $F$ whose $g^\alpha$-potential reproduces that of $\vartheta$ quasi-everywhere on $F$, and the extremal measure $\xi_{F,f}$, the strong limit common to all minimizing sequences. The decisive mechanism is the refined principle of positivity of mass for $\alpha$-Green potentials: from $U^\xi_{g^\alpha}\geqslant U^{\vartheta^F_g}_{g^\alpha}$ quasi-everywhere on $F$ it forces $\xi(D)\geqslant\vartheta^F_g(D)$, and that inequality converts the mass threshold $\vartheta(D)\geqslant1$ into existence of the minimizer.
What would settle it
Choose a domain $D$ whose complement has zero $\alpha$-Riesz capacity, an unbounded relatively closed $F\subset D$ that is not $\alpha$-thin at infinity, and a bounded $\vartheta$ on $D\setminus F$ with $\vartheta(D)>1$ satisfying the distance condition (2.1). For an exhausting family of compact sets $K\subset F$, compute the minimizers $\lambda_{K,f}$ of the compact problems; by Lemma 3.4 these converge strongly and vaguely to the extremal measure $\xi_{F,f}$. The theorem predicts $\xi_{F,f}(D)=1$; displaying such a configuration with $\xi_{F,f}(D)<1$ would refute the equivalence, while the numerical convergence to total mass 1 would confirm it.
Extended reading notes
Core claim
The core claim is Theorem 2.4: fix $\alpha\in(0,n)$ with $\alpha\leqslant2$; if $\alpha=2$, require $\Omega:=D\setminus F$ to be connected; and assume the $\alpha$-harmonic measure condition $\omega_\alpha(x,\mathbb R^n\setminus D;\Omega)=0$ for all $x\in\Omega$ (condition (2.14)). Then the weighted minimum $\alpha$-Green energy problem is solvable if and only if $\vartheta(D)\geqslant1$, and equivalently if and only if $\vartheta^F_g(D)\geqslant1$. In the boundary case $\vartheta(D)=1$, the unique minimizer is $\vartheta^F_g$, the $\alpha$-Green balayage of $\vartheta$ onto $F$, with weighted equilibrium constant $c_{F,f}=0$ and minimum value $-I_{g^\alpha}(\vartheta^F_g)$. The proof routes through the extremal measure $\xi_{F,f}$: a minimizer exists exactly when $\xi(D)=1$, and the sufficiency direction pivots on the refined principle of positivity of mass for $\alpha$-Green potentials, which forces $\xi(D)\geqslant\vartheta^F_g(D)>1$ when $\vartheta^F_g(D)>1$, contradicting the general bound $\xi(D)\leqslant1$.
Load-bearing premise
The load-bearing premise is the refined principle of positivity of mass for $\alpha$-Green potentials, quoted from the author's earlier paper and not re-proved here; if that principle fails for the $\alpha$-Green kernel, the sufficiency direction of the main equivalence ('$\vartheta(D)\geqslant1$ implies existence') collapses.
Editorial extensions
If this is right
- Complete threshold criterion: under the stated hypotheses, existence of the minimizer is equivalent to the single scalar condition $\vartheta(D)\geqslant1$, with no separate finite-capacity assumption on $F$.
- Explicit solution at the threshold: when $\vartheta(D)=1$, the unique minimizer is the $\alpha$-Green balayage $\vartheta^F_g$, with weighted equilibrium constant $c_{F,f}=0$.
- Failure below the threshold: if $\vartheta^F_g(D)<1$, the extremal measure has total mass strictly below 1, so the infimum is not attained by any probability measure on $F$.
- Support description: when the solution exists in the cases covered by the support theorem, its support is the reduced kernel $\check F$ for $\alpha<2$, and $\partial_D\check F$ otherwise.
- Stability: approximating $F$ from inside by compact sets, or from outside by decreasing relatively closed sets, makes the minimizers converge strongly and vaguely to $\lambda_{F,f}$, with the weighted equilibrium constants converging (monotonically under extra balance assumptions).
Reading between the lines
- A parallel mass-threshold criterion is natural in the companion $\alpha$-Riesz problem on unbounded sets, where the paper's reduction indicates that the Riesz balayage mass should replace the Green balayage mass.
- Condition (2.14) is stated pointwise on $\Omega$, while Theorem 1.10 works with $\mu$-a.e. hypotheses; one could try to weaken (2.14) to $\vartheta$-a.e., and if the equivalence survives, the criterion would apply to a broader class of external fields.
- A deliberate check of the excluded case $\alpha=2$ with disconnected $\Omega$ would delimit the theorem: since connectedness enters only through Theorem 1.10, a counterexample there would identify where the boundary condition must be reformulated.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies the Gauss variational (minimum energy with external field) problem for the α-Green kernel g^α_D on a domain D ⊂ R^n, with an external field generated by a fixed bounded measure ϑ concentrated on D \ F, and with admissible probability measures supported on a relatively closed set F. The central result, Theorem 2.4, gives, under condition (2.14) and connectedness of Ω = D \ F unless α < 2, an 'if and only if' criterion for existence of a minimizer: existence holds exactly when ϑ(D) ≥ 1, equivalently when ϑ^F_g(D) ≥ 1; in the boundary case ϑ(D)=1 the minimizer is the α-Green balayage ϑ^F_g. The paper also characterizes the support of the minimizer (Theorem 2.5), establishes convergence of minimizers when F is approximated from inside or outside (Theorems 2.6 and 2.7), and treats the special case c_{κα}(Y)=0 with F not α-thin at infinity (Theorem 2.10). The methods rely on perfectness of the α-Green kernel, α-Riesz and α-Green balayage, and a dual extremal problem with a finite-energy balayage. The paper carefully repairs a gap in a prior paper by Fuglede and Zorii (Remark 1.3 and Lemma 1.7).
Significance. If the main results are correct, the paper provides a complete necessary-and-sufficient existence criterion for a large class of weighted α-Green energy problems, thereby going substantially beyond earlier work where only sufficient conditions or compact-set arguments were available. The paper also answers, for this setting, Ohtsuka's open question about the support of the minimizer. The author is careful with assumptions, explicitly identifies a repaired gap in [15], and gives precise statements of the new theorems. The main caveat is that a decisive step in the proof of Theorem 2.4 depends on a 'refined principle of positivity of mass for g^α-potentials' quoted from the author's in-press paper [30, Theorem 5.1] and not stated or proved here. Because that step is load-bearing for the sufficiency half of the main equivalence, the manuscript is not yet fully self-contained at the point where its central claim is established.
major comments (2)
- [Section 6, proof of Theorem 2.4] The sufficiency half for the case ϑ^F_g(D) > 1 hinges on the 'refined principle of positivity of mass for g^α-potentials' quoted from [30, Theorem 5.1]. After assuming C_ξ ≥ 0, the proof derives U^ξ_g ≥ U^{ϑ^F_g}_g q.e. on F and then invokes that principle to conclude ξ(D) ≥ ϑ^F_g(D) > 1, contradicting (3.14). The ordinary positivity-of-mass principle cited at (1.19) applies only to a measure and its balayage, not to two arbitrary finite-energy measures ξ and ϑ^F_g supported on F with an inequality of their potentials q.e. on F. Since this step is the only one forcing C_ξ < 0, and hence the only one establishing existence when ϑ(D) > 1, the proof is incomplete as written unless the precise statement of [30, Theorem 5.1] is included and it is verified that ξ and ϑ^F_g satisfy all of its hypotheses, especially any regularity, concentration, or extra connectedness or non-thinness conditions beyond (2.14).
- [Section 1.7 and Section 6, proof of Theorem 2.4] Theorem 1.10 is stated as a 'slight improvement' of [30, Theorem 3.7] and its proof is delegated rather than given. This theorem is used essentially in the proof of Theorem 2.4 to conclude ϑ(D) = ϑ^F_g(D) under condition (2.14), and again in the proof of Theorem 2.10. Since [30] is in press and not part of the present manuscript, the exact statement of the cited theorem and a verification that its hypotheses hold for the measure ϑ of this paper (for instance, the condition that ϑ|_F is c_{κα}-absolutely continuous, which is trivial here since ϑ is concentrated on Ω, but should be stated) need to be supplied or the proof of Theorem 1.10 should be included.
minor comments (4)
- [Section 3.2, proof of Lemma 3.6] In the proof of Lemma 3.6, several occurrences of the symbol 'A' (for example 'K ↑ A', 'K ∈ C_A', and 'as K ↑ A') appear to be typos for 'F'; please correct them.
- [Section 8, proof of Theorem 2.6] The same typo 'K ∈ C_A' appears in the proof of Theorem 2.6 after convergence of the net (λ_{K,f}); it should be 'K ∈ C_F'.
- [Section 6, proof of Theorem 2.4] The symbol q is used for the total mass q := ϑ^F_g(D) in the proof of Theorem 2.4, although q was already used in Section 1.1 as a normalization constant for measure classes; consider renaming one of them to avoid confusion.
- [Theorem 2.4] The phrase 'λ_{F,f} exists ⇐⇒ ϑ(D) ≥ 1, or equivalently λ_{F,f} exists ⇐⇒ ϑ^F_g(D) ≥ 1' is logically compressed; stating the equivalence as a single chain 'λ_{F,f} exists ⇐⇒ ϑ(D) ≥ 1 ⇐⇒ ϑ^F_g(D) ≥ 1' would be clearer.
Circularity Check
No significant circularity: Theorem 2.4's equivalence is derived from independent potential-theoretic input, not from its own statement.
full rationale
The paper's central result, Theorem 2.4, is not forced by its own definitions or by a fitted parameter renamed as a prediction. The sufficiency proof in Section 6 uses the refined principle of positivity of mass for α-Green potentials as stated in [30, Theorem 5.1]; this is a load-bearing self-citation, but it is a statement about comparing total masses of two α-Green potentials under q.e. potential ordering, not a restatement of the existence criterion being proved. The cited theorem is external to the present derivation chain and is not shown to include the target result as an assumption. The other main tools—the perfectness of the α-Green kernel, the balayage identities (1.17)–(1.19), and the completeness of E_g^+(F,1) under finite capacity—are prior results with independent content, applied rather than assumed in the desired conclusion. In the case ϑ^F_g(D)<1, unsolvability is established by explicitly constructing minimizing measures and identifying the extremal measure as ϑ^F_g, which is not circular. In the case ϑ^F_g(D)>1, the proof derives a contradiction from the cited positivity-of-mass principle and then converts the resulting sign condition into unit total mass of the extremal measure; again no equation is presupposed in the form of the conclusion. No step exhibits the required reduction 'Eq. X = Eq. Y by construction', so no circularity is present. The heavy self-citation pattern is a transparency and verification issue, but it does not make the derivation circular.
Assumptions & free parameters
assumptions (6)
- domain assumption The α-Green kernel g^α_D is perfect (Fuglede and Zorii [15]).
- standard math The α-Riesz kernel is strictly positive definite.
- standard math The Frostman maximum principle and the domination principle hold for the α-Riesz and α-Green kernels.
- domain assumption Unless α < 2, the set Ω = D \ F is connected.
- domain assumption The α-harmonic measure condition ω_α(x, R^n \ D; Ω) = 0 on all of Ω (condition (2.14)).
- domain assumption The refined principle of positivity of mass for α-Green potentials [30, Theorem 5.1] holds.
Cite this review
Pith. "Pith review of Weighted minimum $\alpha$-Green energy problems." pith.science (2026). https://pith.science/paper/ZJACCKXO
@misc{pith2026250502260,
author = {Pith},
title = {Pith review of: Weighted minimum $\alpha$-Green energy problems},
year = {2026},
howpublished = {\url{https://pith.science/paper/ZJACCKXO}},
note = {Machine review of arXiv:2505.02260}
}
abstract
For the $\alpha$-Green kernel $g^\alpha_D$ on a domain $D\subset\mathbb R^n$, $n\geqslant2$, associated with the $\alpha$-Riesz kernel $|x-y|^{\alpha-n}$, where $\alpha\in(0,n)$ and $\alpha\leqslant2$, and a relatively closed set $F\subset D$, we investigate the problem on minimizing the Gauss functional \[\int g^\alpha_D(x,y)\,d(\mu\otimes\mu)(x,y)-2\int g^\alpha_D(x,y)\,d(\vartheta\otimes\mu)(x,y),\] $\vartheta$ being a given positive (Radon) measure concentrated on $D\setminus F$, and $\mu$ ranging over all probability measures of finite energy, supported in $D$ by $F$. For suitable $\vartheta$, we find necessary and/or sufficient conditions for the existence of the solution to the problem, give a description of its support, provide various alternative characterizations, and prove convergence theorems when $F$ is approximated by partially ordered families of sets. The analysis performed is substantially based on the perfectness of the $\alpha$-Green kernel, discovered by Fuglede and Zorii (Ann. Acad. Sci. Fenn. Math., 2018).
Forward citations
Cited by 1 Pith paper
-
Wasserstein gradient flows of Maximum Mean Discrepancy with energy kernels
Global well-posedness, particle mean-field limits, rigidity of critical points, and obstructions to uniform convergence rates are established for Wasserstein gradient flows of squared MMD with energy kernels -|x|^q, 0...
Reference graph
Works this paper leans on
-
[30]
Zorii N.: Balayage, equilibrium measure, and Deny’s pr inciple of positivity of mass for α -Green potentials. Anal. Math. Phys. 15:3 (2025)
work page 2025
-
[15]
Fuglede, B., Zorii, N.: Green kernels associated with R iesz kernels. Ann. Acad. Sci. Fenn. Math. 43, 121–145 (2018)
work page 2018
-
[1]
Armitage, D.H., Gardiner, S.J.: Classical Potential Th eory. Springer, Berlin (2001)
work page 2001
-
[2]
An Analyti c and Probabilistic Approach to Bal- ayage
Bliedtner, J., Hansen, W.: Potential Theory. An Analyti c and Probabilistic Approach to Bal- ayage. Springer, Berlin (1986)
work page 1986
-
[3]
Bogdan, K., Byczkowski, T., Kulczycki, T., Ryznar, M., S ong, R., Vondraček, Z.: Potential Analysis of Stable Processes and its Extensions. Springer, Berlin (2009)
work page 2009
- [4]
-
[5]
Bourbaki, N.: General Topology. Chapters 5–10. Springe r, Berlin (1989)
work page 1989
- [6]
Show all 31 references
-
[7]
Les Cours Sorbonne, Paris (1961)
Brelot, M.: Eléments de la Théorie Classique du Potentia l. Les Cours Sorbonne, Paris (1961)
1961
-
[8]
Springer, Berlin (1971)
Brelot, M.: On Topologies and Boundaries in Potential Th eory. Springer, Berlin (1971)
1971
-
[9]
Acta Math
Deny, J.: Les potentiels d’énergie finie. Acta Math. 82, 107–183 (1950)
1950
-
[10]
Springer, Berlin (1984)
Doob, J.L.: Classical Potential Theory and Its Probabi listic Counterpart. Springer, Berlin (1984)
1984
-
[11]
Theory and Applic ations
Edwards, R.E.: Functional Analysis. Theory and Applic ations. Holt, Rinehart and Winston, New York (1965)
1965
-
[12]
Frostman, O.: Sur les fonctions surharmoniques d’ordr e fractionnaire. Ark. Mat. 26 A , no. 16 (1939)
1939
-
[13]
Acta Math
Fuglede, B.: On the theory of potentials in locally comp act spaces. Acta Math. 103, 139–215 (1960)
1960
-
[14]
Fuglede, B.: The quasi topology associated with a count ably subadditive set function. Ann. Inst. Fourier Grenoble 21, 123–169 (1971)
1971
-
[16]
Hiroshima Math
Kurokawa, T., Mizuta, Y.: On the order at infinity of Ries z potentials. Hiroshima Math. J. 9, 533–545 (1979)
1979
-
[17]
Springer, Berlin (1972)
Landkof, N.S.: Foundations of Modern Potential Theory . Springer, Berlin (1972)
1972
-
[18]
Ohtsuka, M.: On potentials in locally compact spaces. J . Sci. Hiroshima Univ. Ser. A-I 25, 135–352 (1961)
1961
-
[19]
Acta Szeged 9, 1–42 (1938)
Riesz, M.: Intégrales de Riemann–Liouville et potenti els. Acta Szeged 9, 1–42 (1938)
1938
-
[20]
Grundlehren der mathema- tischen Wissenschaften, 316
Saff, E.B., Totik, V.: Logarithmic Potentials with Exte rnal Fields. Grundlehren der mathema- tischen Wissenschaften, 316. Springer (2024)
2024
-
[21]
Ukrainian Math
Zorii, N.V.: Equilibrium potentials with external fiel ds. Ukrainian Math. J. 55, 1423–1444 (2003)
2003
-
[22]
Zorii, N.: A theory of inner Riesz balayage and its appli cations. Bull. Pol. Acad. Sci. Math. 68, 41–67 (2020)
2020
-
[23]
Potential Anal
Zorii, N.: Harmonic measure, equilibrium measure, and thinness at infinity in the theory of Riesz potentials. Potential Anal. 57, 447–472 (2022)
2022
-
[24]
Zorii, N.: Balayage of measures on a locally compact spa ce. Anal. Math. 48, 249–277 (2022)
2022
-
[25]
Potential Anal
Zorii, N.: On the theory of capacities on locally compac t spaces and its interaction with the theory of balayage. Potential Anal. 59, 1345–1379 (2023)
2023
-
[26]
Potential Anal
Zorii, N.: On the theory of balayage on locally compact s paces. Potential Anal. 59, 1727–1744 (2023)
2023
-
[27]
Zorii, N.: Minimum Riesz energy problems with external fields. J. Math. Anal. Appl. 526, 127235 (2023) Weighted minimum α -Green energy problems 21
2023
-
[28]
Zorii, N.: Minimum energy problems with external fields on locally compact spaces. Constr. Approx. 59, 385–417 (2024)
2024
-
[29]
Zorii N.: On an extension of Fuglede’s theory of pseudo- balayage and its applications. Expo. Math. (2024), 125630, https://doi.org/10.1016/j.exmath .2024.125630
2024
-
[31]
Zorii N.: Inner Riesz balayage in minimum energy proble ms with external fields. Constr. Ap- prox., to appear, arXiv:2306.12788v2 Institute of Mathematics, Academy of Sciences of Ukraine, Ter eshchenkivska 3, 02000, Kyiv, Ukraine, natalia.zorii@gmail.com
Reviewed August 16, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.