Multiplicity obstruction with bounded time-polynomial Bézout certificates
OpenWeierstrassEllipticZeta.bounded_bezout_multiplicity_obstructionLet be a complex period pair with lattice , invariants , normalized entire sigma differential data , and entire functions without a common zero, satisfying
Let be a -linear map equal to the canonical quasiperiods. Put
Let be the locus
Let
Suppose a bijection has nonzero homogeneous coordinate vectors
for all . Assume also that is equipped with a projective additive-fiber model : the projection has exactly the Weierstrass cubic as image, the action is
it obeys the additive action laws, and if and only if there is a unique with . Suppose respects the additive inclusion :
Write for a chosen nonzero representative of the underlying projective point .
Assume in addition the following polynomial differential equations for the normalized coordinate functions.
On the open set where , put
Then these functions have complex derivatives
On the open set where , put
Their complex derivatives are
Each assertion holds at every point where its specified denominator is nonzero, without excluding lattice points.
Assume also the following algebraic description of the chart jets.
For complex parameters , consider the two polynomial derivations on four-variable polynomial rings over :
in variables , and
in variables . Their respective cubic polynomials are
Given five functions , the coordinate maps are
used on and , respectively. These are the affine coordinates for the zero and second homogeneous-coordinate charts of the elliptic-extension model.
For , the derivation annihilates its cubic:
For every polynomial in the four coordinates and every integer ,
At each , the exact derivative identity is
and the vanishing-order criterion is
The analytic order is allowed to be infinite. These statements include and the zero polynomial, using degree zero for the zero polynomial.
Fix also a nondecreasing function with positive values such that, for every degree bound , chart , polynomial of degree at most , and point , vanishing of for all is equivalent to vanishing for all . This one cutoff function is fixed before the polynomial and finite-set parameters below.
For either chart, a point and an integer , define the contact ideal
Angle brackets mean the ideal generated by the displayed set. The definition uses the ambient four-variable polynomial ring and an arbitrary affine evaluation point. Neither the cubic equation nor a nonzero chart denominator is required.
Assume the following proved contact-ideal structure.
Write for the maximal ideal of the affine point . For the fixed invariants of , both charts, all affine points and all nonnegative integers , the following hold:
- A polynomial belongs to if and only if its jets of orders vanish at . Thus taking the ideal span in the definition adds no further polynomials.
- , and .
- If , then and is a primary ideal. Here primary means that the ideal is proper and that in the ideal and outside it imply that a power of lies in the ideal.
- If , then is the entire polynomial ring, including when either order is zero.
- .
- For every finite set , orders and assigned polynomials , there is one polynomial satisfying for every . Equivalently, simultaneously matches the jets of the assigned polynomials through order at each point.
The last assertion is interpolation of polynomial residue classes; no degree bound on the interpolating polynomial is asserted. The statement includes empty finite sets and zero orders. These properties give the local algebra of the contact conditions; the additional dimension hypotheses below describe their combined quotient.
There is a real constant uniform in integers , finite sets containing zero, and complex bihomogeneous polynomials of bidegree in two additive and five projective coordinates. Assume that for every chart and point ,
The normalized polynomial is defined as follows.
Write a polynomial in seven variables as . Define two algebra homomorphisms into four-variable polynomial rings over by
These substitutions set the additive homogenizing coordinate to one and eliminate a projective coordinate using . The subscripts denote the nonzero homogeneous coordinate, so the Lean indices 0 and 1 select and , respectively.
Suppose for every and ,
For each chart , define the finite sets
and the ideal
Assume for each chart that , the quotient is finite-dimensional over , and
For each chart put and define the time polynomial
Assume that is monic of degree and that there are polynomials , all of degree below , with
Assume also the exact membership criterion, for every ,
Degree here assigns minus infinity to zero. An empty chart set has , and , so the assertion includes it. These generator and divisibility hypotheses replace the previous existence-and-uniqueness normal-form clause.
For the same coordinate polynomials, write . Assume the following exact calculation after imposing any additional polynomial equation. For every , set and . There exists a complex algebra isomorphism
The quotient is finite-dimensional and satisfies
with the further bound whenever . This includes an identically zero added equation and empty chart sets.
Assume also that each of the above intersection lengths decomposes into local contact multiplicities. For every chart and every added equation , there exist natural numbers for , with , such that for ,
and
The are the local vanishing orders truncated at the prescribed contact order . All previously assumed quotient isomorphisms, finite dimensionality, gcd-degree formulas, degree bounds, triangular generators and membership criteria continue to hold for the same coordinate polynomials. This added identity is supplied by the proved local-multiplicity lemma; it places no additional restriction on the point set or the added equation.
Assume in addition that the derivatives below the existing cutoff have a finite Bézout certificate in each contact quotient. Put and . For every chart there are coefficient polynomials with
and
This certificate follows from the existing nonzero-jet hypothesis, the containment of powers of point-evaluation kernels in contact ideals, and their finite Chinese remainder property. All previous finite lengths, triangular presentations, quotient isomorphisms, degree bounds and sums of local multiplicities remain in force.
Assume that the finite derivative certificate can be chosen with bounded coefficients in the time coordinate. For each chart put and . There are univariate polynomials , , such that
where and natural subtraction is truncated at zero. For , the residual polynomial
belongs to and satisfies
This is supplied by reducing the earlier coefficient polynomials modulo the same monic in the same time presentation. Every earlier certificate, unit-ideal equality, finite length, triangular presentation, intersection isomorphism and local multiplicity sum remains in force. The coefficient bound depends on the contact length ; an estimate independent of the contact data is not asserted.
The factor counts the actual complex points in that chart: the first coordinate of is , so is injective. Empty chart sets are allowed. Under these hypotheses,
there are an additive subgroup and nonnegative integers with , such that either and , or and , and
Here , , and is the additive quotient map. The triple-sumset vanishing is expressed as membership in an intersection of contact ideals, together with its exact complex quotient dimension. The map takes values in the part of the explicit projective algebraic locus covered by the two standard affine charts with coordinate indices zero and two. The explicit additive action and its fibers are supplied, and the parametrization is bijective and equivariant, and its derivatives are supplied in both projective charts. Identification with an algebraic-group embedding, the geometric multiplicity bound and subgroup projection profiles remain open obligations. The supplied bijection is between the underlying sets; topological and algebraic compatibility remain to be proved. No finiteness of subgroup intersections is assumed.
import Mathlib.RingTheory.AdjoinRoot import Mathlib.Algebra.Polynomial.AlgebraMap import Mathlib.LinearAlgebra.FiniteDimensional.Defs import Mathlib.RingTheory.Ideal.Quotient.Basic import Definitions.Def_WeierstrassEllipticZeta_ProjectiveContactIdeal import Mathlib.RingTheory.Ideal.IsPrimary import Definitions.Def_WeierstrassEllipticZeta_ProjectiveChartNormalization import Definitions.Def_WeierstrassEllipticZeta_ProjectiveChartCalculus import Definitions.Def_WeierstrassEllipticZeta_ProjectiveExtensionFiberModel import Definitions.Def_WeierstrassEllipticZeta_ProjectiveExtensionChartLocus import Definitions.Def_WeierstrassEllipticZeta_ProjectiveExtensionLocus import Mathlib.LinearAlgebra.Projectivization.Basic import Definitions.Def_TranscendenceTheory_GraphQuotientExtension import Mathlib.Analysis.Analytic.Order import Definitions.Def_WeierstrassEllipticZeta_SigmaDifferential import Mathlib.Algebra.MvPolynomial.Eval import Mathlib.Analysis.Calculus.IteratedDeriv.Defs import Mathlib.Algebra.Group.Pointwise.Finset.Basic import Mathlib.LinearAlgebra.Quotient.Basic import Mathlib.Data.Set.Card open WeierstrassEllipticZeta open scoped Pointwise open TranscendenceTheory open scoped Classical
theorem WeierstrassEllipticZeta.bounded_bezout_multiplicity_obstruction
(L : PeriodPair) (D : EllipticSigmaDifferentialData L)
(S : Fin 5 → ℂ → ℂ)
(hS : ∀ j, AnalyticOnNhd ℂ (S j) Set.univ)
(hS_value : ∀ z : ℂ, z ∉ L.lattice → ∀ j : Fin 5,
S j z = D.sigma z ^ 3 * ![1, L.weierstrassP z, L.derivWeierstrassP z,
weierstrassZeta L z,
L.derivWeierstrassP z * weierstrassZeta L z + 2 * L.weierstrassP z ^ 2] j)
(hS_ne : ∀ z : ℂ, ∃ j : Fin 5, S j z ≠ 0)
(η : L.lattice →ₗ[ℤ] ℂ)
(hη : ∀ ω : L.lattice, η ω = zetaQuasiPeriod L ω)
(P : GraphQuotientExtension L.lattice η ≃ ProjectiveExtensionChartLocus L.g₂ L.g₃)
(hP : ∀ z u : ℂ, ∃ hv :
![S 0 z, S 1 z, S 2 z, S 3 z + u * S 0 z, S 4 z + u * S 2 z] ≠ 0,
(P ((extensionPeriodGraph L.lattice η).mkQ (z, u))).val.val = Projectivization.mk ℂ
![S 0 z, S 1 z, S 2 z, S 3 z + u * S 0 z, S 4 z + u * S 2 z] hv)
(F : ProjectiveExtensionFiberModel L.g₂ L.g₃)
(hP_action : ∀ (u : ℂ) (e : GraphQuotientExtension L.lattice η),
P (e + extensionInclusion L.lattice η u) = F.action u (P e))
(hflow :
(∀ z : ℂ, S 0 z ≠ 0 →
HasDerivAt (fun w => S 1 w / S 0 w) (S 2 z / S 0 z) z ∧
HasDerivAt (fun w => S 2 w / S 0 w) (6 * (S 1 z / S 0 z) ^ 2 - L.g₂ / 2) z ∧
HasDerivAt (fun w => S 3 w / S 0 w) (-S 1 z / S 0 z) z) ∧
(∀ z : ℂ, S 2 z ≠ 0 →
HasDerivAt (fun w => S 0 w / S 2 w)
(-6 * (S 1 z / S 2 z) ^ 2 + L.g₂ / 2 * (S 0 z / S 2 z) ^ 2) z ∧
HasDerivAt (fun w => S 1 w / S 2 w)
(-(1 / 2 : ℂ) - L.g₂ * (S 0 z / S 2 z) * (S 1 z / S 2 z) -
3 * L.g₃ / 2 * (S 0 z / S 2 z) ^ 2) z ∧
HasDerivAt (fun w => S 4 w / S 2 w)
(-2 * L.g₂ * (S 1 z / S 2 z) ^ 2 -
3 * L.g₃ * (S 0 z / S 2 z) * (S 1 z / S 2 z)) z))
(hjets :
(∀ c : Fin 2, extensionChartDerivation L.g₂ L.g₃ c (extensionChartCubic L.g₂ L.g₃ c) = 0) ∧
∀ (c : Fin 2) (p : MvPolynomial (Fin 4) ℂ) (n : ℕ),
((extensionChartDerivation L.g₂ L.g₃ c)^[n] p).totalDegree ≤ p.totalDegree + n ∧
∀ z : ℂ, S (extensionChartDenominator c) z ≠ 0 →
iteratedDeriv n (fun w => MvPolynomial.eval (extensionChartCoordinates S c w) p) z =
MvPolynomial.eval (extensionChartCoordinates S c z)
((extensionChartDerivation L.g₂ L.g₃ c)^[n] p) ∧
((n : ℕ∞) ≤ analyticOrderAt
(fun w => MvPolynomial.eval (extensionChartCoordinates S c w) p) z ↔
∀ k < n, MvPolynomial.eval (extensionChartCoordinates S c z)
((extensionChartDerivation L.g₂ L.g₃ c)^[k] p) = 0))
(B : ℕ → ℕ)
(hB : Monotone B ∧ (∀ d : ℕ, 0 < B d) ∧
∀ (d : ℕ) (c : Fin 2) (p : MvPolynomial (Fin 4) ℂ), p.totalDegree ≤ d →
∀ v : Fin 4 → ℂ,
((∀ k < B d, MvPolynomial.eval v ((extensionChartDerivation L.g₂ L.g₃ c)^[k] p) = 0) ↔
∀ k : ℕ, MvPolynomial.eval v ((extensionChartDerivation L.g₂ L.g₃ c)^[k] p) = 0))
(hcontact :
(∀ (c : Fin 2) (v : Fin 4 → ℂ) (n : ℕ) (p : MvPolynomial (Fin 4) ℂ),
p ∈ extensionChartContactIdeal L.g₂ L.g₃ c v n ↔
∀ k < n, MvPolynomial.eval v ((extensionChartDerivation L.g₂ L.g₃ c)^[k] p) = 0) ∧
(∀ (c : Fin 2) (v : Fin 4 → ℂ) (m n : ℕ),
extensionChartContactIdeal L.g₂ L.g₃ c v m * extensionChartContactIdeal L.g₂ L.g₃ c v n ≤
extensionChartContactIdeal L.g₂ L.g₃ c v (m + n)) ∧
(∀ (c : Fin 2) (v : Fin 4 → ℂ) (n : ℕ),
RingHom.ker (MvPolynomial.eval v) ^ n ≤ extensionChartContactIdeal L.g₂ L.g₃ c v n) ∧
(∀ (c : Fin 2) (v : Fin 4 → ℂ) (n : ℕ), 0 < n →
(extensionChartContactIdeal L.g₂ L.g₃ c v n).radical = RingHom.ker (MvPolynomial.eval v) ∧
(extensionChartContactIdeal L.g₂ L.g₃ c v n).IsPrimary) ∧
(∀ (c : Fin 2) (v w : Fin 4 → ℂ), v ≠ w → ∀ m n : ℕ,
extensionChartContactIdeal L.g₂ L.g₃ c v m ⊔ extensionChartContactIdeal L.g₂ L.g₃ c w n = ⊤) ∧
(∀ (c : Fin 2) (v : Fin 4 → ℂ) (n : ℕ) (p : MvPolynomial (Fin 4) ℂ),
p ∈ extensionChartContactIdeal L.g₂ L.g₃ c v (n + 1) →
extensionChartDerivation L.g₂ L.g₃ c p ∈ extensionChartContactIdeal L.g₂ L.g₃ c v n) ∧
(∀ (c : Fin 2) (V : Finset (Fin 4 → ℂ)) (n : V → ℕ)
(p : V → MvPolynomial (Fin 4) ℂ),
∃ q : MvPolynomial (Fin 4) ℂ, ∀ v : V,
q - p v ∈ extensionChartContactIdeal L.g₂ L.g₃ c v.val (n v))) :
∃ C : ℝ, 0 < C ∧ ∀ m n U : ℕ,
1 ≤ m → 1 ≤ n → 1 ≤ U → ∀ X : Finset ℂ,
0 ∈ X → ∀ Q : MvPolynomial (Fin 7) ℂ,
(∀ d ∈ Q.support, d 0 + d 1 = m ∧
d 2 + d 3 + d 4 + d 5 + d 6 = n) →
(∀ (c : Fin 2) (z : ℂ), S (extensionChartDenominator c) z ≠ 0 →
extensionChartNormalize c Q ∉ extensionChartContactIdeal L.g₂ L.g₃ c
(extensionChartCoordinates S c z) (B (m + 2 * n))) →
(∀ (c : Fin 2) (k : ℕ),
((extensionChartDerivation L.g₂ L.g₃ c)^[k] (extensionChartNormalize c Q)).totalDegree ≤
m + 2 * n + k) →
(∀ c : Fin 2,
let Z := (X + X + X).filter (fun z => S (extensionChartDenominator c) z ≠ 0)
let V := Z.image (extensionChartCoordinates S c)
let I : Ideal (MvPolynomial (Fin 4) ℂ) :=
⨅ v : V, extensionChartContactIdeal L.g₂ L.g₃ c v.val (3 * U + 1)
extensionChartNormalize c Q ∈ I ∧
(∃ a : Fin (B (m + 2 * n)) → MvPolynomial (Fin 4) ℂ,
1 - ∑ k : Fin (B (m + 2 * n)), a k *
((extensionChartDerivation L.g₂ L.g₃ c)^[k.val] (extensionChartNormalize c Q)) ∈ I ∧
I ⊔ Ideal.span (Set.range (fun k : Fin (B (m + 2 * n)) =>
(extensionChartDerivation L.g₂ L.g₃ c)^[k.val] (extensionChartNormalize c Q))) = ⊤) ∧
FiniteDimensional ℂ (MvPolynomial (Fin 4) ℂ ⧸ I) ∧
Module.finrank ℂ (MvPolynomial (Fin 4) ℂ ⧸ I) = (3 * U + 1) * Z.card ∧
let M : Polynomial ℂ :=
∏ v : V, (Polynomial.X - Polynomial.C (v.val 0)) ^ (3 * U + 1)
M.Monic ∧ M.degree = ((3 * U + 1) * Z.card : ℕ) ∧
∃ r : Fin 3 → Polynomial ℂ,
(∀ i, (r i).degree < ((3 * U + 1) * Z.card : ℕ)) ∧
I = Ideal.span (insert (Polynomial.aeval (MvPolynomial.X (0 : Fin 4)) M)
(Set.range (fun i : Fin 3 => MvPolynomial.X i.succ -
Polynomial.aeval (MvPolynomial.X (0 : Fin 4)) (r i)))) ∧
(∀ p : MvPolynomial (Fin 4) ℂ,
p ∈ I ↔ M ∣ MvPolynomial.aeval (Fin.cons Polynomial.X r) p) ∧
(∃ b : Fin (B (m + 2 * n)) → Polynomial ℂ,
(∀ k : Fin (B (m + 2 * n)),
(b k).degree < ((3 * U + 1) * Z.card : ℕ) ∧
(Polynomial.aeval (MvPolynomial.X (R := ℂ) (0 : Fin 4)) (b k)).totalDegree ≤
(3 * U + 1) * Z.card - 1) ∧
(1 - ∑ k : Fin (B (m + 2 * n)),
Polynomial.aeval (MvPolynomial.X (0 : Fin 4)) (b k) *
((extensionChartDerivation L.g₂ L.g₃ c)^[k.val] (extensionChartNormalize c Q))) ∈ I ∧
(1 - ∑ k : Fin (B (m + 2 * n)),
Polynomial.aeval (MvPolynomial.X (0 : Fin 4)) (b k) *
((extensionChartDerivation L.g₂ L.g₃ c)^[k.val] (extensionChartNormalize c Q))).totalDegree ≤
((3 * U + 1) * Z.card - 1) + (m + 2 * n + (B (m + 2 * n) - 1))) ∧
∀ p : MvPolynomial (Fin 4) ℂ,
let q := MvPolynomial.aeval (Fin.cons Polynomial.X r) p
let J := I ⊔ Ideal.span {p}
Nonempty ((MvPolynomial (Fin 4) ℂ ⧸ J) ≃ₐ[ℂ]
(Polynomial ℂ ⧸ Ideal.span {gcd M q})) ∧
FiniteDimensional ℂ (MvPolynomial (Fin 4) ℂ ⧸ J) ∧
Module.finrank ℂ (MvPolynomial (Fin 4) ℂ ⧸ J) = (gcd M q).natDegree ∧
Module.finrank ℂ (MvPolynomial (Fin 4) ℂ ⧸ J) ≤ M.natDegree ∧
(q ≠ 0 → Module.finrank ℂ (MvPolynomial (Fin 4) ℂ ⧸ J) ≤ q.natDegree) ∧
∃ e : V → ℕ,
(∀ v : V, e v ≤ 3 * U + 1 ∧ ∀ k ≤ 3 * U + 1,
k ≤ e v ↔ ∀ j < k,
MvPolynomial.eval v.val
((extensionChartDerivation L.g₂ L.g₃ c)^[j] p) = 0) ∧
Module.finrank ℂ (MvPolynomial (Fin 4) ℂ ⧸ J) = ∑ v : V, e v) →
∃ H : Submodule ℤ (GraphExtensionGroup L.lattice η), ∃ a b : ℕ,
((a = 1 ∧ H ≤ LinearMap.ker (extensionAdditiveProjection L.lattice η)) ∨
(a = 0 ∧ H ≤ LinearMap.ker (extensionEllipticProjection L.lattice η))) ∧
b ≤ 2 ∧
((U + 1 : ℕ) : ℝ) *
(H.mkQ '' (extensionCurve L.lattice η '' (X : Set ℂ))).ncard ≤
C * (m : ℝ) ^ a * (n : ℝ) ^ b := by sorry