REVIEW 5 major objections 5 minor 40 references
Choquet-type extension theory of set-pair functions, and applications to graph limits, hypergraphs, Riemannian manifolds and metric measure spaces
T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read Choquet-type extension theory yields a monotonicity inequality for global extension constants that unifies Cheeger, spectral gap, torsional rigidity, and isoperimetric bounds across graph limits, hypergraphs, Riemannian manifolds, and…
desk verdict The core Choquet-extension theory is clean and novel, but the advertised applications lean on unproved identifications and the author's unpublished preprints; send to a serious referee but expect heavy revision. 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 central object is the global extension constant $c(\Phi_p, J\cdot K_q) = \inf_{Y \in \mathcal{Y}} \sup_{f \in Y} \Phi_p(f) / JfK_q$, where $\Phi_p$ is the $L^p$ integral over a parameter space of Choquet extensions $b\varphi_e(f)$ of a family of set functions, and $JfK_q$ is the $L^q$ norm of an intrinsic infinity norm $|f|_{\varphi_e}$. The disjoint-pair version $\Psi_p$ uses Choquet extensions $b\psi_e(f)$ of set-pair functions, which handle signed structures. The proof of the monotonicity inequality runs through the Mazur map $f^r(x) = |f(x)|^r \operatorname{sign}(f(x))$: the substitution turns an $L^p$ norm of $b\varphi_e(f^t)$ into an $L^{ps}$ norm of $b\varphi_e(f)$ times an $L^{qt}$ norm of $|f|$, and the admissible-set structure of the constants converts the pointwise inequality into the global one.
What would settle it
Compute both sides of equation (10) for $p=1$ on a simple two-block constant graphon: if the functional infimum over nonconstant functions differs from the set-based conductance $\inf_S \operatorname{cut}(S)/\min(\operatorname{vol}(S), \operatorname{vol}(J\setminus S))$, the identification in Example 5.1 fails. Alternatively, search finite graphs for a violation of the $p=1$, $r=2$ inequality $h^2/4 \le \lambda_2 \le 2h$; any violation would disprove Theorem 5.2 and hence the monotonicity statement it is derived from.
Extended reading notes
Core claim
On its own terms, the paper claims that a single family of constants controls a wide range of combinatorial and geometric parameters. The load-bearing theorem is the Extended Monotonicity Inequality: whenever $p, q, s, t \ge 1$ satisfy $ps' = qt'$ for Hölder conjugates, one has $c(\Phi_p, J\cdot K_q) \le t\cdot c(\Phi_{ps}, J\cdot K_{qt})$, and $p\cdot c(\Phi_p, J\cdot K_p)$ is nondecreasing in $p$. Under additional degree-related and concentration conditions, Theorem 4.12 gives the two-sided strengthening $(2C_\Phi)^{1-t} c^t(\Phi_{pt}, \widehat{J\cdot K}_{qt}) \le c(\Phi_p, \widehat{J\cdot K}_q) \le t\cdot c(\Phi_{ps}, \widehat{J\cdot K}_{qt})$, and the same statements hold for the disjoint-pair version built from set-pair functions. The paper then identifies these constants with conductance, second Laplacian eigenvalues, maxcut, bipartiteness ratio, hypergraph expansion, torsional rigidity, and p-Laplacian eigenvalues, deriving the corresponding inequalities in its applications section.
Load-bearing premise
The load-bearing step is the identification of abstract global extension constants with concrete quantities: in Example 5.1 the intrinsic norm $|f|_{\varphi_{xy}} = \max(|f(x)|, |f(y)|)$ is replaced by the $p$-mean $((|f(x)|^p + |f(y)|^p)/2)^{1/p}$, and the resulting $c_2(\Phi_1, \widehat{J\cdot K}_1)$ is declared to be the graphon conductance; if that replacement changes the infimum, the Cheeger and spectral corollaries do not follow from the theory.
Editorial extensions
If this is right
- For $L^2$-graphons and graphings, the $p=1$, $r=2$ case gives $h_W^2/4 \le \lambda_2(L_W) \le 2h_W$, covering the known Cheeger inequalities for graph limits.
- For oriented hypergraphs, the machinery yields $2^{p-1} h^p / p^p \le \lambda_2(\Delta_p) \le 2^{p-1} h$, a Cheeger inequality for hypergraph $p$-Laplacians.
- On compact Riemannian manifolds, the same inequality implies that $p\cdot c_k(\|\nabla\cdot\|_p, \|\cdot\|_p)$ increases with $p$, and that $T_p(M)\lambda_p(M) \le \operatorname{vol}(M)^{p-1}$, a Pólya–Szegő type bound on torsional rigidity.
- On metric measure spaces, the paper derives $p \lambda_{k,p}(X)^{1/p} \le q \lambda_{k,q}(X)^{1/q}$ for $1 \le p < q < \infty$.
- Spectral bounds for maxcut and bipartiteness ratio on graphons follow, including $4\operatorname{MaxCut}(W) \le \|W\|_1 \lambda_{\max}(L_W)$ and $\beta_W^2 \le 4 - \lambda_{\max}(L_W) \le 4\beta_W$.
Reading between the lines
- A natural test is whether the constants $2$ and $t$ appearing in the bounds are sharp; the degree-related structure of Theorem 4.12 may admit refinements depending on the degree distribution of the graphon or hypergraph.
- The functional reformulations in Section 3 make fractional set optimizations accessible to convex analysis, so one could construct minimizing sequences and algorithms for maxcut, bipartiteness ratio, and conductance directly from the Rayleigh-type quotients.
- If the identification in Example 5.1 is accepted, the same monotonicity machinery should transfer to other limit objects, such as signed graphons and limits of delta matroids, since the disjoint-pair version is built for exactly those structures.
- The paper leaves implicit the equality cases of Theorem 4.12; identifying the functions that realize the infimum would give extremal eigenfunctions for $p$-Laplacians across different values of $p$.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper develops a Choquet-type extension theory for set-pair functions, introduces L^p-integrated Choquet extensions Phi_p and Psi_p with associated global extension constants c(Phi_p, J.K_q), and proves a monotonicity inequality (Theorem 4.4, and its sharpened version Theorem 4.12 under degree-related and concentration assumptions). The abstract framework is then applied to graphons, graphings, hypergraphs, Riemannian manifolds and metric measure spaces to derive Cheeger-type inequalities, spectral bounds for maxcut and bipartiteness ratio, bounds for hypergraph p-Laplacians, and inequalities involving torsional rigidity and p-Laplacian eigenvalues. The core abstract monotonicity inequality is proved through a Mazur-map argument combined with Holder's inequality, and Theorem 3.2 gives a functional reformulation of set-based fractional optimization via sub- and super-Choquet extensions.
Significance. If the advertised applications are fully justified, the paper would provide a genuinely unifying framework: one monotonicity inequality generating Cheeger, spectral-gap, torsional-rigidity and isoperimetric estimates across very different structures is an appealing and potentially influential contribution. The abstract development, especially the disjoint-pair Choquet extension, the characterization of bisubmodularity by convexity (Theorem 2.9), and the L^p-integration of Choquet integrals, is original and likely to be useful beyond the specific applications. The proofs of Theorems 2.9, 3.2 and the main inequality (8) are largely coherent and checkable from the text; the paper also gives explicit functional representations for maxcut, bipartiteness ratio, conductance, frustration and Dirichlet p-isoperimetric constants. However, several load-bearing identifications between the abstract constants and concrete graphon/manifold quantities are only asserted, with key steps deferred to the author's own unpublished preprints ([38], [39], [40]), and a substantial block of applications in Section 5.4 is stated without proofs.
major comments (5)
- [Section 5.1, Example 5.1 and Eq. (10)] The central identification c_2(Phi_1, fJK_1)=h_W, and similarly c_2(Phi_2, fJK_2)^2=lambda_2(L_W), is asserted in Example 5.1 with the justification "due to Theorem 3.2 and Example 3.6, as well as the kernel reduction lemma in [40]". This is not a consequence of the definitions alone: Theorem 3.2 gives the quotient representation of h_W as an infimum over functions, whereas c_2(Phi_p, fJK_p) is a genus-2 inf-sup over symmetric sets, and the equality of these two quantities is a nontrivial minimax theorem. Moreover, the text replaces the intrinsic norm |f|_{phi_xy}=max(|f(x)|,|f(y)|) by the degree-weighted p-mean norm fJK_p, which is not an equality of norms but a change of the ambient norm used in the definition of c(.,.). Since Corollaries 5.3 and 5.6 and the later spectral bounds depend directly on Eq. (10), this identification is load-bearing and must be proved in the present manuscript, or the kernel reduction lemma of [40] must be stated and proved in sufficient detail for the reader to verify the equality.
- [Sections 4.2 and 6.4, Theorem 4.12 and Propositions 6.7, 6.11, 6.12] The proof of the strengthened monotonicity inequality rests on a lengthy estimate in Proposition 6.7, involving quantities s_{e,i}, I_e and e_phi,f that are introduced in the proof but not fully formalized; several inequalities in the chain are asserted without sufficient justification, and the final passage to the degree-related norm is not fully explained. Furthermore, Proposition 6.12, which is needed for the p=q=1 strengthening in Theorem 4.12, is proved in a single sentence invoking "a statement similar to Theorem 3.2". The theorem also relies on an inequality from [38, Lemma A.1] that is not stated in the paper. Given that Theorem 4.12 is the engine behind the Cheeger inequalities of Section 5.1, the proof must be made fully verifiable or the auxiliary results must be stated with complete proofs.
- [Section 5.4] The text states "We omit the detailed proofs of all the results in this section" immediately before Corollaries 5.22, 5.23 and Proposition 5.24. These results include the torsional-rigidity bound T_p(M) lambda_p(M) <= vol(M)^{p-1} and the lower bounds involving the distance function, which are among the paper's advertised applications. Moreover, the preceding identifications T_p(M)=c_1(nabla, L^1)^{-p} and lambda_p(M)=c_1(nabla, L^p)^p are not derived in the text. Since these results are presented as theorems of the abstract theory rather than as conjectures, the proofs, or at least the precise reduction from the abstract constants to the manifold quantities, should be included.
- [Example 3.6 and Corollary 3.9, item 3] The displayed alternative formula for the graphon conductance, with numerator 2|W|_1|f|_inf - integral W(x,y)|f(x)+f(y)| dxdy, is claimed to be equal to integral W(x,y)|f(x)-f(y)| dxdy. This equality is false in general: for nonnegative f, the alternative numerator equals integral W(x,y)(2|f|_inf - f(x)-f(y)) dxdy, which is generally strictly larger than integral W(x,y)|f(x)-f(y)| dxdy, the latter being integral W(x,y)(2 max(f(x),f(y))-f(x)-f(y)) dxdy. If the formula is intended as an upper bound, or holds only under additional conditions, this must be stated and proved; as written, the displayed equality in Corollary 3.9 item 3 is an internal inconsistency.
- [General (dependence on preprints [38], [39], [40])] Several load-bearing ingredients are cited to the author's own unpublished preprints: the kernel reduction lemma of [40] in Example 5.1, the inequality used in Proposition 6.11 as [38, Lemma A.1], and the (p,q)-Sobolev constant terminology of [39]. For a journal submission, these results should either be restated in the paper or the dependence should be made explicit enough that a reader can verify the claims without access to the preprints. The paper's novelty claim is also harder to assess while the boundary between new results and results imported from these preprints is not drawn precisely.
minor comments (5)
- [Definitions 4.1 and 4.2] The notation for the integrated norms is garbled in the typesetting: the symbols qfy_p and qy_p appear to be intended as L^q-norms of |f|_{phi_e} and |f|_{psi_e}; these should be typeset correctly and defined in words.
- [Notation throughout] The distinction between J.K_p and fJK_p is essential (Theorem 4.4 versus Theorem 4.12), but the notation is not introduced with a dedicated table and the difference is easy to miss; a short notational glossary would substantially improve readability.
- [Section 3.2.3 and Example 3.4] The definitions of psi_2 in Example 3.4 and the computation bpsi_2(f)=2|f|_infty should specify how pairs involving the empty set are treated in the layer-cake integral, to avoid ambiguity about the value at t=0.
- [Section 5.1, Example 5.1] The phrase "this is equivalent to directly using fJK_p instead of JK_p" is imprecise: the p-mean ((|f(x)|^p+|f(y)|^p)/2)^{1/p} is not equal to |f|_{phi_xy}=max(|f(x)|,|f(y)|); it is a different norm, and the equivalence is exactly what needs proof.
- [Section 5.2, Eq. (14)] The expression in Eq. (14) defines c(Phi_p, fJK_q) only after specifying that the numerator uses |f(x)-f(y)|^p with respect to the measure eta induced by W; as written it is easy to misread the exponent of the numerator as missing a factor 1/p.
Circularity Check
Abstract monotonicity theorem is self-contained, but the graphon/graphing applications hinge on an unproved kernel-reduction lemma cited from the author's own preprint.
-
self citation load bearing
[Section 5.1, Example 5.1]
"The Y2-type min-max value c2(Φp, fJ·Kp) unifies many quantities, e.g. the Cheeger constant hW = c2(Φ1, fJ·K1) (due to Theorem 3.2 and Example 3.6, as well as the kernel reduction lemma in [40]), and the second smallest eigenvalue of graphon Laplacian λ2(LW ) = (c2(Φ2, fJ·K2))^2."
This equality is the bridge that turns the abstract monotonicity inequality Theorem 4.12 into the advertised Cheeger inequalities (Corollaries 5.3 and 5.6). The paper does not prove the bridge; it delegates the reduction from the genus-2 min-max constant c2 to the plain Rayleigh quotient hW to the kernel reduction lemma in [40], the author's own unpublished preprint. The statement of the lemma is not given, and the corresponding identification for λ2 is merely asserted. The graphon and graphing Cheeger bounds therefore rest on a load-bearing self-citation rather than on a derivation contained in this paper. The abstract monotonicity theorem itself is proved from the definitions, so the circularity is confined to the application bridge, not to Theorem 4.4 or 4.12.
full rationale
The core theoretical result, Theorem 4.4 (extended monotonicity inequality), is derived in Section 6.3 from the definitions of Choquet-type extensions, L^p integration, and the global extension constants, using Hölder inequalities and the Mazur map. No parameter is fitted, and the inequality is not made true by definition. The strengthened Theorem 4.12 is likewise proved from the degree-related and concentration hypotheses. These central derivations are self-contained and show no circularity. The main circularity concern is in the applications: Example 5.1 identifies c2(Φ1, fJ·K1) with the graphon conductance hW and c2(Φ2, fJ·K2)^2 with λ2(LW), citing the author's unpublished kernel reduction lemma [40] without stating or proving it. Since the Cheeger, spectral-gap, and related corollaries in Section 5 depend on this identification, the advertised unification for graph limits is not fully derived in the paper; it rests on a load-bearing self-citation. This does not make the main monotonicity inequality circular, but it does prevent the paper from being entirely self-contained at the application level. I therefore assign a score of 4: some self-citation is load-bearing, while the central abstract claim retains independent content.
Assumptions & free parameters
assumptions (6)
- domain assumption Finite Borel measure space (J,B,μ) with 0<μ(J)<∞
- domain assumption Admissible set families Y invariant under Mazur maps, specifically Y_k={A⊂B_d\{0}: genus(A)≥k}
- domain assumption Degree-related structure: there exists deg:J→[0,∞) satisfying ∫_E ∑_{x∈e_J} g(x) dη(e)=∫_J deg(x)g(x)dμ(x) for all g∈B_d
- domain assumption Concentration of φ_e on e_J, with |e_J|≤2 and C_Φ=sup_e∥φ_e∥∞<∞
- ad hoc to paper Kernel reduction lemma from [40] (author's own preprint) equating c_2(Φ_1,fJ·K_1) with graphon conductance
- standard math Lemma A.1 from [38] (author's published paper): ||b|^{t-1}b-|a|^{t-1}a| ≥ |b-a| ((|a|^t+|b|^t)/2)^{1-1/t}
invented entities (2)
-
Disjoint-pair Choquet extension bψ
-
L^p-integrated Choquet extensions Φ_p, Ψ_p and global extension constants c(Φ_p,J·K_q)
Cite this review
Pith. "Pith review of Choquet-type extension theory of set-pair functions, and applications to graph limits, hypergraphs, Riemannian manifolds and metric measure spaces." pith.science (2026). https://pith.science/paper/ERVOM35B
@misc{pith2026260809606,
author = {Pith},
title = {Pith review of: Choquet-type extension theory of set-pair functions, and applications to graph limits, hypergraphs, Riemannian manifolds and metric measure spaces},
year = {2026},
howpublished = {\url{https://pith.science/paper/ERVOM35B}},
note = {Machine review of arXiv:2608.09606}
}
abstract
We propose Choquet extension for set-pair functions, $L^p$ integration of Choquet extensions, and global extension constants, and apply these to investigate optimization problems and bound many combinatorial and geometric quantities. Our research line is also applicable to the study of original Choquet extension; within this framework, parallel results for the original version are obtained. Specifically, we use Choquet-type extensions to build an equivalent functional representation of set-based fractional optimization, which finds applications in various settings, such as maxcut, bipartiteness ratio, and conductance on graph limits or Riemannian manifolds. We further establish the $L^p$ integration of a family of Choquet extensions, and apply it to derive spectral bounds for conductance and other combinatorial quantities on measure spaces. A monotonicity inequality on global extension constants is proposed, which unifies classical estimates and uncovers new bounds for a lot of geometric and combinatorial quantities, such as torsional rigidity, Cheeger constants, Dirichlet $p$-isoperimetric constant, and $p$-Laplacian eigenvalues, in totally distinct underlying structures----including hypergraphs, graph limits, Riemannian manifolds and metric measure spaces.
Reference graph
Works this paper leans on
-
[40]
Hidden critical and Morse equivalence behind duality: Theory and Applications
Dong Zhang. Hidden critical and morse equivalence behind duality: Theory and applications, 2026. arXiv:2606.27004. 39
work page Pith review arXiv 2026
-
[38]
Homological eigenvalues of graphp-Laplacians.J
Dong Zhang. Homological eigenvalues of graphp-Laplacians.J. Topol. Anal., 17(2):555–606, 2025
work page 2025
-
[39]
Graphon as a Bridge between Graphs and Manifolds
Dong Zhang. Graphon as a bridge between graphs and manifolds, 2026. arXiv:2607.20213
work page Pith review arXiv 2026
-
[1]
Ziegler.Oriented matroids, volume 46 ofEncyclopedia of Mathematics and its Applications
Anders Bj¨ orner, Michel Las Vergnas, Bernd Sturmfels, Neil White, and G¨ unter M. Ziegler.Oriented matroids, volume 46 ofEncyclopedia of Mathematics and its Applications. Cambridge University Press, Cambridge, second edition, 1999
work page 1999
-
[2]
S´ os, Bal´ azs Szegedy, and Katalin Veszter- gombi
Christian Borgs, Jennifer Chayes, L´ aszl´ o Lov´ asz, Vera T. S´ os, Bal´ azs Szegedy, and Katalin Veszter- gombi. Graph limits and parameter testing. InSTOC’06: Proceedings of the 38th Annual ACM Symposium on Theory of Computing, pages 261–270. ACM, New York, 2006
work page 2006
-
[3]
S´ os, and Katalin Vesztergombi
Christian Borgs, Jennifer Chayes, L´ aszl´ o Lov´ asz, Vera T. S´ os, and Katalin Vesztergombi. Counting graph homomorphisms. InTopics in discrete mathematics, volume 26 ofAlgorithms Combin., pages 315–371. Springer, Berlin, 2006
work page 2006
-
[4]
Chayes, L´ aszl´ o Lov´ asz, Vera T
Christian Borgs, Jennifer T. Chayes, L´ aszl´ o Lov´ asz, Vera T. S´ os, and Katalin Vesztergombi. Con- vergent sequences of dense graphs ii. multiway cuts and statistical physics.Annals of Mathematics, 176(1):151–219, 2012
work page 2012
-
[5]
Greedy algorithm and symmetric matroids.Math
Andr´ e Bouchet. Greedy algorithm and symmetric matroids.Math. Programming, 38(2):147–159, 1987
work page 1987
Show all 40 references
-
[6]
Inequalities between torsional rigidity and principal eigenvalue of thep-Laplacian.Calc
Luca Briani, Giuseppe Buttazzo, and Francesca Prinari. Inequalities between torsional rigidity and principal eigenvalue of thep-Laplacian.Calc. Var. Partial Differential Equations, 61(2):Paper No. 78, 25, 2022
2022
-
[7]
Sharp geometric and functional inequalities in metric measure spaces with lower Ricci curvature bounds.Geom
Fabio Cavalletti and Andrea Mondino. Sharp geometric and functional inequalities in metric measure spaces with lower Ricci curvature bounds.Geom. Topol., 21(1):603–645, 2017
2017
-
[8]
Hubert Chan, Anand Louis, Zhihao Gavin Tang, and Chenzi Zhang
T.-H. Hubert Chan, Anand Louis, Zhihao Gavin Tang, and Chenzi Zhang. Spectral properties of hypergraph Laplacian and approximation algorithms.J. ACM, 65(3):Art. 15, 48, 2018
2018
-
[9]
Chandrasekaran and Santosh N
R. Chandrasekaran and Santosh N. Kabadi. Pseudomatroids.Discrete Math., 71(3):205–217, 1988
1988
-
[10]
Theory of capacities.Ann
Gustave Choquet. Theory of capacities.Ann. Inst. Fourier (Grenoble), 5:131–295, 1953/54
1953
-
[11]
A sharp upper bound for the torsional rigidity of rods by means of web functions.Arch
Graziano Crasta, Ilaria Fragal` a, and Filippo Gazzola. A sharp upper bound for the torsional rigidity of rods by means of web functions.Arch. Ration. Mech. Anal., 164(3):189–211, 2002
2002
-
[12]
The equality case in Cheeger’s and Buser’s inequalities onRCDspaces.J
Nicol` o De Ponti, Andrea Mondino, and Daniele Semola. The equality case in Cheeger’s and Buser’s inequalities onRCDspaces.J. Funct. Anal., 281(3):Paper No. 109022, 36, 2021
2021
-
[13]
Sharp bounds for the first eigenvalue and the torsional rigidity related to some anisotropic operators.Math
Francesco Della Pietra and Nunzia Gavitone. Sharp bounds for the first eigenvalue and the torsional rigidity related to some anisotropic operators.Math. Nachr., 287(2-3):194–209, 2014
2014
-
[14]
Hypergraphp-Laplacians and scale spaces.J
Ariane Fazeny, Daniel Tenbrinck, Kseniia Lukin, and Martin Burger. Hypergraphp-Laplacians and scale spaces.J. Math. Imaging Vision, 66(4):529–549, 2024
2024
-
[15]
Elsevier B
Satoru Fujishige.Submodular functions and optimization, volume 58 ofAnnals of Discrete Mathe- matics. Elsevier B. V., Amsterdam, second edition, 2005
2005
-
[16]
The total varia- tion on hypergraphs - learning on hypergraphs revisited
Matthias Hein, Simon Setzer, Leonardo Jost, and Syama Sundar Rangapuram. The total varia- tion on hypergraphs - learning on hypergraphs revisited. In C.J. Burges, L. Bottou, M. Welling, Z. Ghahramani, and K. Weinberger, editors,Advances in Neural Information Processing Systems...
2013
-
[17]
Skew bisubmodularity and valued CSPs.SIAM J
Anna Huber, Andrei Krokhin, and Robert Powell. Skew bisubmodularity and valued CSPs.SIAM J. Comput., 43(3):1064–1084, 2014
2014
-
[18]
Vietnam Journal of Mathematics, 50(2):323–358, 2022
J¨ urgen Jost, Raffaella Mulas, and Dong Zhang.p-Laplace operators for oriented hypergraphs. Vietnam Journal of Mathematics, 50(2):323–358, 2022. 38
2022
-
[19]
Cambridge Studies in Advanced Mathematics
J¨ urgen Jost, Raffaella Mulas, and Dong Zhang.Spectra of discrete structures. Cambridge Studies in Advanced Mathematics. Cambridge University Press, 2026
2026
-
[20]
Discrete-to-continuous extensions: Lov´ asz extension and Morse theory.Discrete & Computational Geometry, 72(1):49–72, 2024
J¨ urgen Jost and Dong Zhang. Discrete-to-continuous extensions: Lov´ asz extension and Morse theory.Discrete & Computational Geometry, 72(1):49–72, 2024
2024
-
[21]
Cheeger inequalities for graph limits.Ann
Abhishek Khetan and Mahan MJ. Cheeger inequalities for graph limits.Ann. Inst. Fourier (Greno- ble), 74(1):257–305, 2024
2024
-
[22]
Subgraph densities in Markov spaces
D´ avid Kunszenti-Kov´ acs, L´ aszl´ o Lov´ asz, and Bal´ azs Szegedy. Subgraph densities in Markov spaces. Adv. Math., 437:Paper No. 109414, 67, 2024
2024
-
[23]
On nonlinear Rayleigh quotients.Potential Anal., 2(3):199–218, 1993
Peter Lindqvist. On nonlinear Rayleigh quotients.Potential Anal., 2(3):199–218, 1993
1993
-
[24]
Hypergraph markov operators, eigenvalues and approximation algorithms
Anand Louis. Hypergraph markov operators, eigenvalues and approximation algorithms. InPro- ceedings of the Forty-Seventh Annual ACM Symposium on Theory of Computing, STOC ’15, page 713–722, New York, NY, USA, 2015. Association for Computing Machinery
2015
-
[25]
Many sparse cuts via higher eigenvalues
Anand Louis, Prasad Raghavendra, Prasad Tetali, and Santosh Vempala. Many sparse cuts via higher eigenvalues. InSTOC’12—Proceedings of the 2012 ACM Symposium on Theory of Comput- ing, pages 1131–1140. ACM, New York, 2012
2012
-
[26]
Submodular functions and convexity
L´ aszl´ o Lov´ asz. Submodular functions and convexity. InMathematical Programming The State of the Art: Bonn 1982, pages 235–257. Springer, 1983
1982
-
[27]
Subgraph densities in signed graphons and the local Simonovits-Sidorenko conjecture
L´ aszl´ o Lov´ asz. Subgraph densities in signed graphons and the local Simonovits-Sidorenko conjecture. Electron. J. Combin., 18(1):Paper 127, 21, 2011
2011
-
[28]
American Mathematical Society, Providence, RI, 2012
L´ aszl´ o Lov´ asz.Large networks and graph limits, volume 60 ofAmerican Mathematical Society Colloquium Publications. American Mathematical Society, Providence, RI, 2012
2012
-
[29]
Flows on measurable spaces.Geom
L´ aszl´ o Lov´ asz. Flows on measurable spaces.Geom. Funct. Anal., 31(2):402–437, 2021
2021
-
[30]
Choquet extension of non-monotone submodular setfunctions.Acta Math
L´ aszl´ o Lov´ asz. Choquet extension of non-monotone submodular setfunctions.Acta Math. Hungar., 176(1):1–14, 2025
2025
-
[31]
Submodular setfunctions on sigma-algebras, version 2, 2023
L´ aszl´ o Lov´ asz. Submodular setfunctions on sigma-algebras, version 2, 2023. arXiv:2302.04704
2023 arXiv
-
[32]
P´ olya and S
G. P´ olya and S. Szeg¨ o. Isoperimetric inequalities in mathematical physics.Annals of Math. Studies, 27, 1951
1951
-
[33]
Directed submodularity, ditroids and directed submodular flows.Math
Li Qun Qi. Directed submodularity, ditroids and directed submodular flows.Math. Programming, 42(3):579–599, 1988. Submodular optimization
1988
-
[34]
Sabina de Lis and Sergio Segura de Le´ on
Jos´ e C. Sabina de Lis and Sergio Segura de Le´ on. The limit asp→1 of the higher eigenvalues of thep-Laplacian operator−∆ p.Indiana Univ. Math. J., 70(4):1395–1439, 2021
2021
-
[35]
InGeometric science of information
Jo Andersson Stokke, Ronny Bergmann, Martin Hanik, and Christoph von Tycowicz.p-Laplacians for manifold-valued hypergraphs. InGeometric science of information. Part III, volume 16035 of Lecture Notes in Comput. Sci., pages 162–171. Springer, Cham, [2026]©2026
2026
-
[36]
Trevisan
L. Trevisan. Max cut and the smallest eigenvalue.SIAM J. Comput., 41(6):1769–1786, 2012
2012
-
[37]
On relations between principal eigen- value and torsional rigidity.Commun
Michiel van den Berg, Giuseppe Buttazzo, and Aldo Pratelli. On relations between principal eigen- value and torsional rigidity.Commun. Contemp. Math., 23(8):Paper No. 2050093, 28, 2021
2021
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.