Documentation

LeanPool.Superdiffusion.Support.CoarseGraining.QuadraticStability.Integral

Integral #

Ported from scottnarmstrong/CoarseGraining at c7ddd76c08ade64fed1b8d2ca51be14dfee8deb4. Module imports, support namespace and local option directives adapted for Lean Pool.

Quadratic stability, item B′3 (integral form) #

The integral half of Lemma 4.1 (l.quadratic.stability) of the high-moment paper (Armstrong–Kuusi–Loher, to appear): stability of the two quadratic minima under an L∞-comparable perturbation of the coefficient field supported on S.

We assume only the two scalar Euler identities at the single test field Y := Zt − Z (no subspace, no minimization, no existence), and derive the sharpened bound with constant 6K.

Chosen integrability package for (h4) #

We assume IntegrableOn over U of the five real-valued pairing integrands that actually appear in the proof:

This is the minimal explicit list sufficient for every integral split, the Cauchy–Schwarz step, and the localization; each hypothesis is a concrete ℝ-valued IntegrableOn, directly dischargeable by a consumer holding L² minimizers. (The energy Zt·BtZt needs no separate hypothesis: it equals Z·BtZ + 2 Z·BtY + Y·BtY a.e. by symmetry.)

No EuclideanSpace.

Two elementary analytic helpers #

theorem HCPolySupport.integral_sqrt_mul_sqrt_le {α : Type u_1} {m : MeasurableSpace α} {μ : MeasureTheory.Measure α} {f g : α → ℝ} (hf : MeasureTheory.Integrable f μ) (hg : MeasureTheory.Integrable g μ) (hf0 : 0 ≤ᵐ[μ] f) (hg0 : 0 ≤ᵐ[μ] g) :
∫ (x : α), √(f x) * √(g x) ∂μ ≤ √(∫ (x : α), f x ∂μ) * √(∫ (x : α), g x ∂μ)

Integral Cauchy–Schwarz for two nonnegative integrable functions: ∫ √f·√g ≤ √(∫f)·√(∫g).

theorem HCPolySupport.sqrt_mul_sqrt_le_half_add {a b : ℝ} (ha : 0 ≤ a) (hb : 0 ≤ b) :
√a * √b ≤ (a + b) / 2

AM–GM in the form √a·√b ≤ (a+b)/2 for nonnegative reals.

theorem HCPolySupport.le_sq_of_le_mul_sqrt {G C : ℝ} (hG : 0 ≤ G) (hC : 0 ≤ C) (h : G ≤ C * √G) :
G ≤ C ^ 2

If 0 ≤ G and G ≤ C·√G with 0 ≤ C, then G ≤ C².

Data and hypotheses for B′3 #

theorem HCPolySupport.abs_setIntegral_energy_sub_le {d : ℕ} {U S : Set (Vec d)} {B Bt : Vec d → BlockMat d} {Z Zt : Vec d → BlockVec d} {K : ℝ} (hU : MeasurableSet U) (hS : MeasurableSet S) (hSU : S ⊆ U) (hK : 1 ≤ K) (hae : ∀ᵐ (x : Vec d) ∂MeasureTheory.volume.restrict U, IsSymmetricBlockMat (B x) ∧ IsSymmetricBlockMat (Bt x) ∧ (∀ (W : BlockVec d), 0 ≤ blockVecDot W (blockMatVecMul (B x) W)) ∧ (∀ (W : BlockVec d), 0 ≤ blockVecDot W (blockMatVecMul (Bt x) W)) ∧ BlockMatLoewnerLE (Bt x) (K • B x) ∧ BlockMatLoewnerLE (B x) (K • Bt x)) (hagree : ∀ᵐ (x : Vec d) ∂MeasureTheory.volume.restrict (U \ S), B x = Bt x) (hIntBZZ : MeasureTheory.IntegrableOn (fun (x : Vec d) => blockVecDot (Z x) (blockMatVecMul (B x) (Z x))) U MeasureTheory.volume) (hIntBtZZ : MeasureTheory.IntegrableOn (fun (x : Vec d) => blockVecDot (Z x) (blockMatVecMul (Bt x) (Z x))) U MeasureTheory.volume) (hIntBtYY : MeasureTheory.IntegrableOn (fun (x : Vec d) => blockVecDot (Zt x - Z x) (blockMatVecMul (Bt x) (Zt x - Z x))) U MeasureTheory.volume) (hIntBtZY : MeasureTheory.IntegrableOn (fun (x : Vec d) => blockVecDot (Z x) (blockMatVecMul (Bt x) (Zt x - Z x))) U MeasureTheory.volume) (hIntBZY : MeasureTheory.IntegrableOn (fun (x : Vec d) => blockVecDot (Z x) (blockMatVecMul (B x) (Zt x - Z x))) U MeasureTheory.volume) (hEulerB : ∫ (x : Vec d) in U, blockVecDot (Zt x - Z x) (blockMatVecMul (B x) (Z x)) = 0) (hEulerBt : ∫ (x : Vec d) in U, blockVecDot (Zt x - Z x) (blockMatVecMul (Bt x) (Zt x)) = 0) :
|(∫ (x : Vec d) in U, blockVecDot (Zt x) (blockMatVecMul (Bt x) (Zt x))) - ∫ (x : Vec d) in U, blockVecDot (Z x) (blockMatVecMul (B x) (Z x))| ≤ 6 * K * ∫ (x : Vec d) in S, blockVecDot (Z x) (blockMatVecMul (B x) (Z x))

B′3. Stability of the two quadratic minima under an L∞-comparable, S-supported perturbation. Constant 6K.