YukonModule.ProximityPrize.SubmissionLower.BoundaryTailZeroCount.part0
DefinitionYukon_273df68c1ff8e6ab04d115a8better-codes
Source module ProximityPrize.SubmissionLower.BoundaryTailZeroCount.
Definition code
import Definitions.Def_Yukon_043883ce6ed32260483e0ffb
set_option backward.isDefEq.respectTransparency.types false
namespace ProximityPrize.SubmissionLower.RCN199
open scoped Classical BigOperators WithZero
open RCN002 RCN344 RCN341 RCN095 RCN114 RCN295 RCN187 RCN207 RCN064 RCN204 RCN271 RCN257
noncomputable section
set_option autoImplicit false
variable {K : Type} [Field K]
variable {P : Ideal (MvPolynomial (Fin 3) K)} [P.IsPrime]
variable {H G : MvPolynomial (Fin 3) K}
namespace MovingPoleBudget
theorem zero_le_of_doubled_pole [IsAlgClosed K]
(budget : MovingPoleBudget P H G)
(base : SeparableLiteralCoordinate P)
(T : MvPolynomial (Fin 3) K) (denom factor : ℕ) (q : FlagDegree)
(hH : H ∉ P) (hT : T ∉ P)
(hpole : ∀ v : Place K (CoordinateField K P),
2 * poleOrder v.val
(coordinateEvaluation K P T /
(coordinateEvaluation K P H)^denom) ≤
2 * flagPole v.val (coordinate K P) q +
(2 * factor : ℕ) * movingPoleTarget P H G v) :
FiniteRegularZeroSetBound P H T
(budget.weightedCost q + factor * budget.movingCost) := by
apply finite_regular_zero_bound_of_separator K P base H T denom _ hT hH
intro W
have hlocal : ∀ v ∈ W,
(2 * poleOrder v.val
(coordinateEvaluation K P T /
(coordinateEvaluation K P H)^denom) : ℤ) ≤
2 * flagPole v.val (coordinate K P) q +
(2 * factor : ℤ) * movingPoleTarget P H G v := by
intro v hv
exact hpole v
have hsum := Finset.sum_le_sum hlocal
have hflag := budget.sum_flagPole_le q W
have hmove := budget.movingPole W
have hbound :
(2 : ℤ) * (∑ v ∈ W, poleOrder v.val
(coordinateEvaluation K P T /
(coordinateEvaluation K P H)^denom)) ≤
(2 : ℤ) * (budget.weightedCost q + factor * budget.movingCost) := by
calc
_ = ∑ v ∈ W, (2 * poleOrder v.val
(coordinateEvaluation K P T /
(coordinateEvaluation K P H)^denom) : ℤ) := by
simp only [Finset.mul_sum]
_ ≤ ∑ v ∈ W, (2 * flagPole v.val (coordinate K P) q +
(2 * factor : ℤ) * movingPoleTarget P H G v) := hsum
_ = (2 : ℤ) * ((∑ v ∈ W, flagPole v.val (coordinate K P) q) +
(factor : ℤ) * ∑ v ∈ W, movingPoleTarget P H G v) := by
simp only [Finset.sum_add_distrib, ← Finset.mul_sum]
push_cast
ring
_ ≤ (2 : ℤ) * (budget.weightedCost q + factor * budget.movingCost) := by
gcongr
have hfinal :
(∑ v ∈ W, poleOrder v.val
(coordinateEvaluation K P T /
(coordinateEvaluation K P H)^denom) : ℤ) ≤
(budget.weightedCost q + factor * budget.movingCost : ℤ) := by
linarith
simpa only [RCN346.poleOrder, coordinateEvaluation_eq_aeval,
Nat.cast_add, Nat.cast_mul] using hfinal
end MovingPoleBudget
end
end ProximityPrize.SubmissionLower.RCN199
Source