Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Elliptic polynomial sums and the two interpolation estimates

Definition
WeierstrassEllipticZeta_PolynomialInterpolation

by tomasz · Sep 14, 2026 · Mathlib 0df444a (Lean v4.33.1)

complex-analysiselliptic-functionsinterpolation

For a complex period lattice Λ\LambdaΛ, let ℘,ζ,℘′\wp,\zeta,\wp'℘,ζ,℘′ denote its canonical Weierstrass functions. For positive integers d,ld,ld,l and a complex coefficient array p=(pijk)p=(p_{ijk})p=(pijk​), set

P(w)=∑i=0d∑j,k=0lpijkwi℘(w)jζ(w)k.P(w)=\sum_{i=0}^{d}\sum_{j,k=0}^{l}p_{ijk}w^i\wp(w)^j\zeta(w)^k.P(w)=i=0∑d​j,k=0∑l​pijk​wi℘(w)jζ(w)k.

The exponents are bounded separately by (d,l,l)(d,l,l)(d,l,l); they need not attain these bounds. For a complex number vvv put

M1(v,l)=max⁡a,b,c≥0, a+b+c≤5l(1+∣ζ(v)a℘(v)b℘′(v)c∣).M_1(v,l)=\max_{a,b,c\ge0,\ a+b+c\le5l}\left(1+|\zeta(v)^a\wp(v)^b\wp'(v)^c|\right).M1​(v,l)=a,b,c≥0, a+b+c≤5lmax​(1+∣ζ(v)a℘(v)b℘′(v)c∣).

The maximum is finite; its constant monomial contributes 222.

The two interpolation predicates encode the following requirements for a given function σ\sigmaσ and real parameters C,R0C,R_0C,R0​, using c13=Cc_{13}=Cc13​=C for the first predicate and c14=Cc_{14}=Cc14​=C for the second. The predicates themselves impose no normalization on σ\sigmaσ or sign restriction on C,R0C,R_0C,R0​; theorems impose the intended positivity and sigma conditions.

For every positive d,l,Md,l,Md,l,M, every array with ∣pijk∣≤M|p_{ijk}|\le M∣pijk​∣≤M, and every u∈Cu\in\mathbb Cu∈C, the expression

F(z)=σ(z+u)3lP(z+u)F(z)=\sigma(z+u)^{3l}P(z+u)F(z)=σ(z+u)3lP(z+u)

has an entire extension. Choose this extension before the radii and derivative data. If 2<r<R2<r<R2<r<R, R>R0R>R_0R>R0​, ∣u∣<R|u|<R∣u∣<R, ∣v∣<r−2|v|<r-2∣v∣<r−2, and FFF has at least N≥0N\ge0N≥0 zeros counted with multiplicity in ∣z∣<r|z|<r∣z∣<r, then for every integer t≥0t\ge0t≥0,

∣F(t)(v)∣≤t!(d+1)(l+1)2M(2R)dc13R2l(2rR)N.|F^{(t)}(v)|\le t!(d+1)(l+1)^2M(2R)^d c_{13}^{R^2l}\left(\frac{2r}{R}\right)^N.∣F(t)(v)∣≤t!(d+1)(l+1)2M(2R)dc13R2l​(R2r​)N.

For every positive d,l,Md,l,Md,l,M, every array with ∣pijk∣≤M|p_{ijk}|\le M∣pijk​∣≤M, and every v∉Λv\notin\Lambdav∈/Λ, the expression

G(z)=σ(z)15l[2(℘(v)−℘(z))]3lP(z+v)G(z)=\sigma(z)^{15l}[2(\wp(v)-\wp(z))]^{3l}P(z+v)G(z)=σ(z)15l[2(℘(v)−℘(z))]3lP(z+v)

has an entire extension. Choose this extension before the radii and derivative data. If 2<r<R2<r<R2<r<R, R>R0R>R_0R>R0​, ∣v∣<R|v|<R∣v∣<R, ∣u∣<r−2|u|<r-2∣u∣<r−2, and GGG has at least N≥0N\ge0N≥0 zeros counted with multiplicity in ∣z∣<r|z|<r∣z∣<r, then for every integer t≥0t\ge0t≥0,

∣G(t)(u)∣≤t!(d+1)(l+1)2(5l+1)6M(2R)dM1(v,l)c14R2l(2rR)N.|G^{(t)}(u)|\le t!(d+1)(l+1)^2(5l+1)^6M(2R)^dM_1(v,l)c_{14}^{R^2l}\left(\frac{2r}{R}\right)^N.∣G(t)(u)∣≤t!(d+1)(l+1)2(5l+1)6M(2R)dM1​(v,l)c14R2l​(R2r​)N.

Formalization note. A lower bound on the number of zeros is expressed by an arbitrary finite set s⊂{∣z∣<r}s\subset\{|z|<r\}s⊂{∣z∣<r} and arbitrary multiplicities m(a)≥0m(a)\ge0m(a)≥0 with N≤∑a∈sm(a)N\le\sum_{a\in s}m(a)N≤∑a∈s​m(a), together with vanishing of every derivative of order below m(a)m(a)m(a) at each a∈sa\in sa∈s. The estimate holds for every such witness. Empty sets, N=0N=0N=0, t=0t=0t=0, and the zero coefficient array are included. The product identities specify the extensions only where all meromorphic factors are regular: z+u∉Λz+u\notin\Lambdaz+u∈/Λ in the first case, and z,z+v∉Λz,z+v\notin\Lambdaz,z+v∈/Λ in the second. Zeros and derivative evaluation points may lie at excluded points of those product formulas, because they refer to the entire extensions. No condition 2r<R2r<R2r<R is imposed. The constants raised to R2lR^2lR2l use real powers; all other displayed exponents are integers.

The polynomial index set, evaluation, and maximum are also defined when either degree parameter is zero. The height at l=0 is 2. The interpolation predicates only require their conclusions for strictly positive d and l. These lightweight definitions record the statement of Lemma 6; they contain no proof obligations asserted as facts.

Definition code
import Definitions.Def_WeierstrassEllipticZeta_SigmaDifferential
import Definitions.Def_WeierstrassEllipticZeta_EntireRegularization
import Mathlib.Data.Finset.Lattice.Fold
import Mathlib.Analysis.SpecialFunctions.Pow.Real

noncomputable section
open Finset Set
open scoped NNReal

namespace WeierstrassEllipticZeta

/-- The rectangular monomial indices for a polynomial of degrees (d,l,l). -/
abbrev EllipticPolynomialIndex (d l : ℕ) := Fin (d + 1) × Fin (l + 1) × Fin (l + 1)

/-- Evaluation of a rectangular polynomial at (z,wp(z),zeta(z)). -/
def ellipticRectangularPolynomial (L : PeriodPair) (d l : ℕ)
    (p : EllipticPolynomialIndex d l → ℂ) (z : ℂ) : ℂ :=
  ∑ i, p i * z ^ i.1.val * L.weierstrassP z ^ i.2.1.val *
    weierstrassZeta L z ^ i.2.2.val

/-- The finite maximum M1 in the cleared estimate of Lemma 6. -/
def ellipticInterpolationHeight (L : PeriodPair) (v : ℂ) (l : ℕ) : ℝ :=
  ↑((Finset.univ.filter (fun i : Fin (5*l+1) × Fin (5*l+1) × Fin (5*l+1) =>
    i.1.val + i.2.1.val + i.2.2.val ≤ 5*l)).sup (fun i =>
      (1 : ℝ≥0) + ‖weierstrassZeta L v ^ i.1.val *
        L.weierstrassP v ^ i.2.1.val * L.derivWeierstrassP v ^ i.2.2.val‖₊))

/-- Entire extension and zero-count derivative estimate for a shifted polynomial. -/
def SigmaPolynomialInterpolation (L : PeriodPair) (σ : ℂ → ℂ) (C R₀ : ℝ) : Prop :=
  ∀ (d l : ℕ), 0 < d → 0 < l →
    ∀ (p : EllipticPolynomialIndex d l → ℂ) (M : ℝ), 0 < M →
      (∀ i, ‖p i‖ ≤ M) → ∀ u : ℂ,
      ∃ F : ℂ → ℂ, AnalyticOnNhd ℂ F univ ∧
        (∀ z : ℂ, z + u ∉ L.lattice →
          F z = σ (z + u) ^ (3*l) * ellipticRectangularPolynomial L d l p (z+u)) ∧
        ∀ (r R : ℝ), 2 < r → r < R → R₀ < R → ‖u‖ < R →
          ∀ v : ℂ, ‖v‖ < r-2 →
          ∀ (N : ℕ) (s : Finset ℂ) (m : ℂ → ℕ),
            (∀ a ∈ s, ‖a‖ < r) →
            (∀ a ∈ s, ∀ j < m a, iteratedDeriv j F a = 0) →
            N ≤ ∑ a ∈ s, m a → ∀ t : ℕ,
              ‖iteratedDeriv t F v‖ ≤
                t.factorial * (d+1 : ℕ) * (l+1 : ℕ)^2 * M * (2*R)^d *
                  C ^ (R^2 * l) * (2*r/R)^N

/-- Entire extension and zero-count derivative estimate after clearing addition poles. -/
def ClearedSigmaPolynomialInterpolation (L : PeriodPair) (σ : ℂ → ℂ)
    (C R₀ : ℝ) : Prop :=
  ∀ (d l : ℕ), 0 < d → 0 < l →
    ∀ (p : EllipticPolynomialIndex d l → ℂ) (M : ℝ), 0 < M →
      (∀ i, ‖p i‖ ≤ M) → ∀ v : ℂ, v ∉ L.lattice →
      ∃ G : ℂ → ℂ, AnalyticOnNhd ℂ G univ ∧
        (∀ z : ℂ, z ∉ L.lattice → z + v ∉ L.lattice →
          G z = σ z ^ (15*l) * (2*(L.weierstrassP v - L.weierstrassP z)) ^ (3*l) *
            ellipticRectangularPolynomial L d l p (z+v)) ∧
        ∀ (r R : ℝ), 2 < r → r < R → R₀ < R → ‖v‖ < R →
          ∀ u : ℂ, ‖u‖ < r-2 →
          ∀ (N : ℕ) (s : Finset ℂ) (m : ℂ → ℕ),
            (∀ a ∈ s, ‖a‖ < r) →
            (∀ a ∈ s, ∀ j < m a, iteratedDeriv j G a = 0) →
            N ≤ ∑ a ∈ s, m a → ∀ t : ℕ,
              ‖iteratedDeriv t G u‖ ≤
                t.factorial * (d+1 : ℕ) * (l+1 : ℕ)^2 * (5*l+1 : ℕ)^6 * M *
                  (2*R)^d * ellipticInterpolationHeight L v l *
                  C ^ (R^2 * l) * (2*r/R)^N

end WeierstrassEllipticZeta
Source
Senthil Kumar K (2026), Algebraic independence of values of Weierstrass elliptic and zeta functions, https://doi.org/10.1017/S001309152610145X, §4, Lemma 6(i)–(ii), equations (14)–(15) and the proof of Lemma 6.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me