Attained optimal anchor weight and exact scalar budget criterion
ProvedWeierstrassEllipticZeta.optimal_anchor_weight_selectionFor the existing sparse-GCD anchor candidates, let k(a) be the number of classes of X modulo the candidate's period kernel and let d(a) be its additive degree. A point has d(a)=m and weight w(a)=k(a)=|X|. A line or whole fibre has d(a)=0 and weight w(a)=(m+1)*k(a).
The theorem proves that the minimum weight W is attained by the chosen optimalSparseAnchor, is at most |X|, is at least 1 if X is nonempty, and is no larger than any anchor's weight. Every minimizing anchor with d(a)=0 satisfies
k(a) <= floor(|X|/(m+1)).
Moreover, for every natural-number local cost E and real constant C, existence of an anchor satisfying
k(a)E <= C(d(a)+1)*n^2
is equivalent to the single scalar inequality
WE <= C(m+1)*n^2.
The same minimum and chosen minimizer work for all E and C. The result requires no analytic, nonvanishing or homogeneity hypotheses beyond the existing definitions; empty X and zero degrees are allowed, with the lower bound on W conditional on X being nonempty.
The new cutoff restricts optimal zero-additive-degree candidates, not all valid lines or fibres. Previously every quotient class count was only trivially bounded by |X|; an optimal nonpoint choice must now have at most floor(|X|/(m+1)) classes. This is a pruning criterion and exact budget reformulation. Feasibility and the minimum remain classical, and the theorem does not prove the uniform geometric estimate or change the integer-search bound.
import Definitions.Def_WeierstrassEllipticZeta_OptimalAnchorWeight import Mathlib.Tactic noncomputable section open scoped Classical open WeierstrassEllipticZeta
theorem WeierstrassEllipticZeta.optimal_anchor_weight_selection
(Λ : Submodule ℤ ℂ) (η : Λ →ₗ[ℤ] ℂ)
(X : Finset ℂ) (S : Fin 5 → ℂ → ℂ) (Q : MvPolynomial (Fin 7) ℂ)
(m n : ℕ) (K : Set ℂ) (Z : Finset ℂ) :
(minimumAnchorWeight Λ η X S Q m n K Z ≤ X.card ∧
(X.Nonempty → 1 ≤ minimumAnchorWeight Λ η X S Q m n K Z) ∧
sparseAnchorWeight Λ η X S Q m n K Z
(optimalSparseAnchor Λ η X S Q m n K Z) =
minimumAnchorWeight Λ η X S Q m n K Z ∧
(∀ a : SparseGCDAnchorCandidate Λ η X S Q m n K Z,
minimumAnchorWeight Λ η X S Q m n K Z ≤
sparseAnchorWeight Λ η X S Q m n K Z a) ∧
(∀ a : SparseGCDAnchorCandidate Λ η X S Q m n K Z,
sparseAnchorWeight Λ η X S Q m n K Z a =
minimumAnchorWeight Λ η X S Q m n K Z →
elementaryDegree (candidateLocusShape Λ η X
(sparseGCDAnchorLocus Λ η X S Q m n K Z a)) m = 0 →
candidateClassCount Λ η X (sparseGCDAnchorLocus Λ η X S Q m n K Z a) ≤
X.card / (m + 1))) ∧
∀ (E : ℕ) (C : ℝ),
(∃ a : SparseGCDAnchorCandidate Λ η X S Q m n K Z,
(((X.image (elementaryPeriodKernel Λ η (candidateLocusShape Λ η X
(sparseGCDAnchorLocus Λ η X S Q m n K Z a))).mkQ).card * E : ℕ) : ℝ) ≤
C * (((elementaryDegree (candidateLocusShape Λ η X
(sparseGCDAnchorLocus Λ η X S Q m n K Z a)) m + 1) * n ^ 2 : ℕ) : ℝ)) ↔
(((minimumAnchorWeight Λ η X S Q m n K Z * E : ℕ) : ℝ) ≤
C * (((m + 1) * n ^ 2 : ℕ) : ℝ)) := by sorry