Duality for Delsarte's extremal problem on compact Gelfand pairs
Pith reviewed 2026-05-15 11:58 UTC · model grok-4.3
The pith
Delsarte extremal problems on compact Gelfand pairs admit strong duality when cast as infinite-dimensional linear programs.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
We study Delsarte-type problems for positive definite functions on compact Gelfand pairs as infinite-dimensional linear programming problems. This setup includes, as a particular case, the case of compact Abelian groups. Depending on the restriction on the signs of the functions, we obtain two important particular cases, the Turán and Delsarte problems. In this paper, we describe their duals and prove a strong duality statement.
What carries the argument
Infinite-dimensional linear programming formulation of extremal problems for positive definite functions on compact Gelfand pairs, together with the explicit dual program and the strong duality theorem that equates their optimal values.
If this is right
- The common optimal value can be computed or bounded by solving either the primal or the dual linear program.
- Sphere-packing and Turán-type bounds on Gelfand pairs become accessible through the dual formulation.
- The same duality recovers classical results for compact Abelian groups as a special case.
- Sign-restricted versions of the problem (Turán versus Delsarte) inherit the same dual structure.
Where Pith is reading between the lines
- The duality statement may serve as a template for extending the linear-programming approach to other classes of symmetric spaces where positive-definite functions are well understood.
- Numerical solution of the dual programs could produce new explicit bounds for packing densities on specific Gelfand pairs.
- The framework suggests examining whether weak duality alone suffices for certain applications when strong duality fails.
Load-bearing premise
The standard strong duality theorem for infinite-dimensional linear programs applies directly to the space of positive definite functions on compact Gelfand pairs.
What would settle it
A concrete compact Gelfand pair together with a feasible positive definite function whose objective value strictly exceeds the optimal value of the explicitly described dual program.
read the original abstract
We study Delsarte-type problems for positive definite functions on compact Gelfand pairs as infinite-dimensional linear programming problems. This setup includes, as a particular case, the case of compact Abelian groups. Depending on the restriction on the signs of the functions, we obtain two important particular cases, the Tur\'an and Delsarte problems. These problems have been studied in relation to number theory, sphere packing, and statistics. In this paper, we describe their duals and prove a strong duality statement.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper formulates Delsarte-type and Turán-type extremal problems for positive definite functions on compact Gelfand pairs (including compact Abelian groups) as infinite-dimensional linear programs, explicitly describes the corresponding dual programs, and proves a strong duality theorem relating the primal and dual values.
Significance. If the strong duality holds under the stated conditions, the work supplies a clean dual description that could streamline the derivation of bounds in sphere packing, coding theory, and statistical applications on Gelfand pairs. The explicit duals also open the possibility of new analytic or numerical attacks on these classical problems.
major comments (1)
- [Section 3 (Duality theorem and proof)] The proof of strong duality (the central claim) invokes a standard infinite-dimensional LP duality theorem but does not explicitly verify the required constraint qualification—e.g., existence of a strictly feasible point in the interior of the positive-definite cone or weak*-compactness of the feasible set—in the uniform topology on C(G). Without this check, a duality gap cannot be ruled out, and the dual description loses its guaranteed optimality property.
minor comments (1)
- [Section 2] Notation for the cone of positive-definite functions and the dual pairing should be introduced once and used consistently; the current alternation between P and the space of measures is occasionally ambiguous.
Simulated Author's Rebuttal
We thank the referee for the careful reading and for highlighting the need for explicit verification of the constraint qualification in the duality theorem. We will revise the manuscript accordingly to strengthen the presentation.
read point-by-point responses
-
Referee: [Section 3 (Duality theorem and proof)] The proof of strong duality (the central claim) invokes a standard infinite-dimensional LP duality theorem but does not explicitly verify the required constraint qualification—e.g., existence of a strictly feasible point in the interior of the positive-definite cone or weak*-compactness of the feasible set—in the uniform topology on C(G). Without this check, a duality gap cannot be ruled out, and the dual description loses its guaranteed optimality property.
Authors: We agree that an explicit check strengthens the argument. In the revised manuscript we will add a dedicated paragraph (or short subsection) verifying the hypotheses of the invoked infinite-dimensional LP duality theorem. Weak*-compactness of the feasible set follows from Alaoglu’s theorem: the unit ball of C(G)* is weak*-compact, and the positive-definite cone intersected with the linear constraints is a weak*-closed subset because G is compact and the positive-definiteness condition is weak*-closed. Strict feasibility holds by taking the constant function f ≡ 1, whose Fourier transform is a positive multiple of the delta at the trivial representation; this function lies in the interior of the positive-definite cone in the uniform norm and satisfies all inequality constraints strictly. Consequently the duality gap is zero and the dual attains its optimum. revision: yes
Circularity Check
No circularity detected in duality derivation
full rationale
The paper formulates Delsarte and Turán problems as infinite-dimensional linear programs over positive definite functions on compact Gelfand pairs and invokes standard strong duality results from functional analysis to obtain the dual problems. This relies on external theorems for infinite-dimensional LPs rather than any self-definitional reduction, fitted parameters renamed as predictions, or load-bearing self-citations. The derivation chain is self-contained against external benchmarks in convex analysis and harmonic analysis on groups, with no steps that reduce by construction to the paper's own inputs or prior author work.
Axiom & Free-Parameter Ledger
axioms (1)
- domain assumption Positive definite functions on compact Gelfand pairs can be treated as variables in an infinite-dimensional linear program with sign restrictions yielding the Turán and Delsarte cases.
Lean theorems connected to this paper
-
IndisputableMonolith/Cost/FunctionalEquation.leanwashburn_uniqueness_aczel unclear?
unclearRelation between the paper passage and the cited Recognition theorem.
We study Delsarte-type problems for positive definite functions on compact Gelfand pairs as infinite-dimensional linear programming problems... prove a strong duality statement.
-
IndisputableMonolith/Foundation/RealityFromDistinction.leanreality_from_one_distinction unclear?
unclearRelation between the paper passage and the cited Recognition theorem.
The dual cone Q* of Q = M ∩ L ... is the cone M* + L*.
What do these tags mean?
- matches
- The paper's claim is directly supported by a theorem in the formal canon.
- supports
- The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
- extends
- The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
- uses
- The paper appears to rely on the theorem as machinery.
- contradicts
- The paper's claim conflicts with a theorem or certificate in the canon.
- unclear
- Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.
Forward citations
Cited by 1 Pith paper
-
Duality for Delsarte's extremal problem on locally compact Abelian groups
Strong duality is proved for a generalized Delsarte extremal problem on locally compact Abelian groups, unifying prior results for finite groups and R^d.
Reference graph
Works this paper leans on
-
[1]
V. V. Arestov and A. G. Babenko,On the Delsarte scheme for estimating contact numbers, Proc. Steklov Inst. Math.4(1997), 36–65
work page 1997
-
[2]
V. V. Arestov and E. E. Berdysheva,Tur´ an’s problem for positive definite functions with supports in a hexagon, Proc. Steklov Inst. Math., Supp 1. (2001), 20–29
work page 2001
-
[3]
V. V. Arestov and E. E. Berdysheva,Tur´ an’s problem for a class of polytopes, East J. Approx. (2002), no. 8, 381–388. 22 BERDYSHEV A AT AL
work page 2002
-
[4]
E. E. Berdysheva and Sz. Gy. R´ ev´ esz,Delsarte’s extremal problem and packing on locally compact Abelian groups, Annali della Scuola Normale di Pisa – Classe di Scienze (5),XXIV(2023), 1007– 1052
work page 2023
-
[5]
E. E. Berdysheva, M. D. Ramabulana, Sz. Gy. R´ ev´ esz,On extremal problems of Delsarte type for positive definite functions on LCA groups, Expo. Math.44(2026), 125663
work page 2026
-
[6]
C. Berg, A. P. Peron, E. Porcu,Orthogonal expansions related to compact Gelfand pairs, Expo. Math. 36(2018), 259–277
work page 2018
-
[7]
H. Brezis,Functional Analysis, Sobolev Spaces, and Partial Differential Equations, Springer-Verlag, New York (2010)
work page 2010
-
[8]
D. L. Cohn,Measure Theory, 2nd edition, Birkh¨ auser, 2013
work page 2013
-
[9]
H. Cohn and N. Elkies,New upper bounds for sphere packings, I, Ann. of Math. (2)157(2003), 689–714
work page 2003
- [10]
-
[11]
Cohn,New upper bounds on sphere packings
H. Cohn,New upper bounds on sphere packings. II, Geom. Topol.6(2002) 329–353
work page 2002
-
[12]
H. Cohn, A. Kumar, S. D. Miller, D. Radchenko and M. Viazovska,The sphere packing problem in dimension 24, Ann. of Math.185(2017) 1017–1033
work page 2017
-
[13]
H. Cohn, Y. Zhao,Sphere packing via spherical codes, Duke Math. J.163(2014), 1965–2002
work page 2014
-
[14]
Delsarte,Bounds for unrestricted codes by linear programming, Philips Res
P. Delsarte,Bounds for unrestricted codes by linear programming, Philips Res. Rep.2(1972), 272–289
work page 1972
-
[15]
P. Delsarte, J.-M. Goethals, and J. J. Seidel,Spherical codes and designs, Geom. Dedicata(3)6 (1977), 363–388
work page 1977
-
[16]
R. J. Duffin,Infinite programs, In: Linear Inequalities and Related Systems (AM-38), ed. by H. W. Kuhn and A. W. Tucker, Princeton University Press, 1956, pp. 157–70
work page 1956
-
[17]
R. E. Edwards,Functional Analysis: Theory and Applications, Reinhart and Winston, New York, 1965
work page 1965
-
[18]
M. Ga´ al and Zs. Nagy-Csiha,On the existence of an extremal function in the Delsarte extremal problem, Mediterr. J. Math.17(2020), no. 190
work page 2020
-
[19]
J.-P. Gabardo,The Tur´ an Problem and Its Dual for Positive Definite Functions Supported on a Ball inR d, J. Fourier Anal. Appl.30(2024), 11
work page 2024
-
[20]
T. Gneiting,Supplement to Strictly and non-strictly positive definite functions on spheres(2013), doi: DOI:10.3150/12-BEJSP06SUPP
-
[21]
E. G. Gol’stein,Teoriya dvoistvennosti v matematicheskom programmirovanii i eyo prilozheniya(The- ory of Duality in Mathematical Programming and Its Applications), Moscow: Nauka, 1971
work page 1971
-
[22]
D. V. Gorbachev,Extremal problems for entire functions of exponential spherical type, connected with the Levenshtein bound on the sphere packing density inR n (Russian), Izvestiya of the Tula State University, Ser. Mathematics, Mechanics, Informatics6(2000) 71–78
work page 2000
-
[23]
D. V. Gorbachev,An extremal problem for periodic functions with supports in the ball, Math. Notes 3(2001), 313–319
work page 2001
-
[24]
Husain,Introduction to Topological Groups, W
T. Husain,Introduction to Topological Groups, W. B. Saunders Company, Philadelphia and London, 1966
work page 1966
-
[25]
A. D. Ioffe, V. M. Tikhomirov,Duality of convex functions and extremum problems, Russian Math. Surveys23(6) (1968), 53–124
work page 1968
-
[26]
G. A. Kabatyanskii, V. I. Levenshtein,On bounds for packing on the sphere and in space(Russian), Probl. Inform.14(1978), no. 1, 3–25
work page 1978
- [27]
-
[28]
M. N. Kolountzakis, M. Matolcsi, and M. Weiner,An application of positive definite functions to the problems of MUBs, Proc. Amer. Math. Soc.146(2018), no. 3, 1143–1150
work page 2018
-
[29]
M. N. Kolountzakis and Sz. Gy. R´ ev´ esz,On a problem of Tur´ an about positive definite functions, Proc. Amer. Math. Soc.131(2003), 3423–3430
work page 2003
-
[30]
N. A. Kuklin,Delsarte method in the problem on kissing numbers in high-dimensional spaces, Proc. Steklov Inst. Math. Suppl.284(2014), 108–123
work page 2014
-
[31]
V. I. Levenshtein,Bounds for packings inn-dimensional Euclidean space, Dokl. Akad. Nauk SSSR 245(1979) 1299–1303
work page 1979
-
[32]
M. Matolcsi, I. Z. Ruzsa,Difference Sets and Positive Exponential Sums I. General Properties, J. Fourier Anal. Appl.20(2014), 17–41
work page 2014
-
[33]
M. Matolcsi and M. Weiner,An Improvement on the Delsarte-Type LP-Bound with Application to MUBs, Open Syst. Inf. Dyn.22(2015), 1550001. DUALITY FOR DELSARTE’S EXTREMAL PROBLEM ON COMPACT GELF AND PAIRS 23
work page 2015
-
[34]
M. D. Ramabulana,On the existence of an extremal function for the Delsarte extremal problem, Anal. Math.51(2025), 279–291
work page 2025
-
[35]
M. D. Ramabulana,Delsarte-type problems on homogeneous spaces and convolutions roots, J. Fourier Anal. Appl.32(2026), 44
work page 2026
-
[36]
Sz. Gy. R´ ev´ esz,Some trigonometric extremal problems and duality, J. Aust. Math. Soc. Ser. A50 (1991), 384–390
work page 1991
-
[37]
Sz. Gy. R´ ev´ esz,On Beurling’s Prime Number Theorem, Period. Math. Hung.28(1994) no. 3, 195– 210
work page 1994
-
[38]
Sz. Gy. R´ ev´ esz,On some extremal problems of Landau, Serdica Math. J.33(1) (2007) 125–162
work page 2007
-
[39]
Sz. Gy. R´ ev´ esz,Tur´ an’s extremal problem on locally compact Abelian groups, Anal. Math.37(2009), 15–50
work page 2009
-
[40]
Richmond,General Topology, De Gruyter, Berlin/Boston, 2020
T. Richmond,General Topology, De Gruyter, Berlin/Boston, 2020
work page 2020
-
[41]
I. Z. Ruzsa,Connections between the uniform distribution of a sequence and its differences, in: Topics in Classical Number Theory, Colloq. Math. Soc. J´ anos Bolyai, vol. 34, North-Holland, Amsterdam– New York–Budapest, 1981, pp. 1419–1443
work page 1981
-
[42]
Sasv´ ari,Positive Definite and Definitizable Functions, Vol
Z. Sasv´ ari,Positive Definite and Definitizable Functions, Vol. 2, Akademie Verlag, Berlin, 1994
work page 1994
-
[43]
C. L. Siegel, ¨Uber Gitterpunkte in konvexen K¨ orpern und ein damit zusammenh¨ angendes Extremal- problem, Acta Math.65(1935), 307–323
work page 1935
-
[44]
M. S. Viazovska,The sphere packing problem in dimension 8, Ann. of Math. (3)185(2017), 991–1015
work page 2017
-
[45]
Virosztek,Applications of an intersection formula to dual cones, Bull
D. Virosztek,Applications of an intersection formula to dual cones, Bull. Aust. Math. Soc.97(2018), 94–101
work page 2018
-
[46]
M. Wackenhuth,Linear programming bounds on homogeneous spaces, I: optimal packing density, ArXiv preprint, arXiv:2505.23572v1
-
[47]
Wackenhuth,Bounds on hyperbolic sphere packings: on a conjecture by Cohn and Zhao, C
M. Wackenhuth,Bounds on hyperbolic sphere packings: on a conjecture by Cohn and Zhao, C. R., Math., Acad. Sci. Paris364(2026), 237–242
work page 2026
-
[48]
J. A. Wolf,Harmonic Analysis on Commutative Spaces, American Mathematical Society, Providence, 2007
work page 2007
-
[49]
V. A. Yudin,Packings of balls in Euclidean space, and extremal problems for trigonometric poly- nomials(Russian), Diskret. Mat.1(1989) 155–158; translation in Discrete Math. Appl.1(1991) 69–72. Elena E. Berdysheva University of Cape Town, South Africa Email address:elena.berdysheva@uct.ac.za B´alint Farkas University of Wuppertal Gaußstraße 20, 42119 Wupp...
work page 1989
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.