fpvandoorn/carleson
Source indexedlemma · leanprover/lean4:v4.32.0
TileStructure.Forest.e764_postCS
Carleson.ForestOperator.RemainingTiles · Carleson/ForestOperator/RemainingTiles.lean:627 to 695
Source documentation
Equation (7.6.4) of Lemma 7.6.2 (after applying Cauchy–Schwarz and simplification).
Exact Lean statement
lemma e764_postCS (hu₁ : u₁ ∈ t) (hu₂ : u₂ ∈ t) (hu : u₁ ≠ u₂) (h2u : 𝓘 u₁ ≤ 𝓘 u₂)
(hf : BoundedCompactSupport f) :
eLpNorm (approxOnCube (𝓙₆ t u₁) (‖adjointCarlesonSum (t u₂ \ 𝔖₀ t u₁ u₂) f ·‖)) 2 volume ≤
C2_1_3 a * 2 ^ (11 * a + 2) *
(∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), (D : ℝ≥0∞) ^ (-k * κ / 2)) *
eLpNorm ((𝓘 u₁ : Set X).indicator (maximalFunction volume 𝓑 c𝓑 r𝓑 1 f ·)) 2 volumeComplete declaration
Lean source
Full Lean sourceLean 4
lemma e764_postCS (hu₁ : u₁ ∈ t) (hu₂ : u₂ ∈ t) (hu : u₁ ≠ u₂) (h2u : 𝓘 u₁ ≤ 𝓘 u₂) (hf : BoundedCompactSupport f) : eLpNorm (approxOnCube (𝓙₆ t u₁) (‖adjointCarlesonSum (t u₂ \ 𝔖₀ t u₁ u₂) f ·‖)) 2 volume ≤ C2_1_3 a * 2 ^ (11 * a + 2) * (∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), (D : ℝ≥0∞) ^ (-k * κ / 2)) * eLpNorm ((𝓘 u₁ : Set X).indicator (maximalFunction volume 𝓑 c𝓑 r𝓑 1 f ·)) 2 volume := by have aem_MB : AEMeasurable (maximalFunction volume 𝓑 c𝓑 r𝓑 1 f) volume := measurable_maximalFunction.aemeasurable classical calc _ ≤ _ := e764_preCS hu₁ hu₂ hu h2u hf _ = C2_1_3 a * 2 ^ (4 * a) * ∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), (∑ J ∈ (𝓙₆ t u₁).toFinset, (volume (J : Set X))⁻¹ * (∫⁻ y in J, maximalFunction volume 𝓑 c𝓑 r𝓑 1 f y * ∑ I with s I = s J - k ∧ Disjoint (I : Set X) (𝓘 u₁) ∧ ¬Disjoint ↑J (ball (c I) (8 * D ^ s I)), (ball (c I) (8 * D ^ s I)).indicator 1 y) ^ 2) ^ (2 : ℝ)⁻¹ := by congr! with k mk J mJ rw [← lintegral_finsetSum']; swap · fun_prop (discharger := measurability) congr with y; rw [mul_comm, Finset.sum_mul] _ ≤ C2_1_3 a * 2 ^ (4 * a) * ∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), (∑ J ∈ (𝓙₆ t u₁).toFinset, (∫⁻ y in J, maximalFunction volume 𝓑 c𝓑 r𝓑 1 f y ^ 2) * (⨍⁻ y in J, (∑ I : Grid X with s I = s J - k ∧ Disjoint (I : Set X) (𝓘 u₁) ∧ ¬Disjoint ↑J (ball (c I) (8 * D ^ s I)), (ball (c I) (8 * D ^ s I)).indicator 1 y) ^ 2 ∂volume)) ^ (2 : ℝ)⁻¹ := by gcongr _ * ∑ k ∈ _, (∑ J ∈ _, ?_) ^ _ with k mk J mJ rw [setLAverage_eq, ENNReal.div_eq_inv_mul, ← mul_assoc, mul_comm _ _⁻¹, mul_assoc] gcongr; apply ENNReal.sq_lintegral_mul_le_mul_lintegral_sq aem_MB.restrict -- Cauchy–Schwarz fun_prop (discharger := measurability) _ ≤ C2_1_3 a * 2 ^ (4 * a) * ∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), (∑ J ∈ (𝓙₆ t u₁).toFinset, (∫⁻ y in J, maximalFunction volume 𝓑 c𝓑 r𝓑 1 f y ^ 2) * C7_6_4 a k) ^ (2 : ℝ)⁻¹ := by gcongr with k mk J mJ; rw [mem_toFinset] at mJ; exact square_function_count mJ _ ≤ C2_1_3 a * 2 ^ (4 * a) * ∑ k ∈ Finset.Icc ⌊C7_6_3 a n⌋ (2 * S), 2 ^ (7 * a + 2) * D ^ (-k * κ / 2) * (∑ J ∈ (𝓙₆ t u₁).toFinset, ∫⁻ y in J, maximalFunction volume 𝓑 c𝓑 r𝓑 1 f y ^ 2) ^ (2 : ℝ)⁻¹ := by gcongr with k mk rw [← Finset.sum_mul, mul_comm _ (C7_6_4 a k : ℝ≥0∞), ENNReal.mul_rpow_of_nonneg _ _ (by positivity)] gcongr rw [C7_6_4, NNReal.mul_rpow, show (8 : ℝ≥0) = 2 ^ (3 : ℝ) by norm_num, ← NNReal.rpow_mul, ← mul_assoc, ← NNReal.rpow_intCast, ← NNReal.rpow_mul, ENNReal.rpow_ofNNReal (by positivity), NNReal.mul_rpow, ← NNReal.rpow_mul, ← NNReal.rpow_add two_ne_zero, ← NNReal.rpow_mul, ENNReal.coe_mul, ENNReal.coe_rpow_of_ne_zero two_ne_zero, ← show (2 : ℝ≥0∞) = (2 : ℝ≥0) by rfl, ENNReal.coe_rpow_of_ne_zero (by norm_cast; unfold defaultD; positivity), show ((D : ℝ≥0) : ℝ≥0∞) = (D : ℝ≥0∞) by rfl, Int.cast_neg, div_eq_mul_inv, ← ENNReal.rpow_natCast] gcongr · exact one_le_two · rw [add_assoc, add_mul, Nat.cast_add, Nat.cast_mul, show (14 * a * 2⁻¹ : ℝ) = 7 * a by ring, Nat.cast_ofNat] gcongr calc _ ≤ (1 + 3 * 1) * (2 : ℝ)⁻¹ := by gcongr; exact κ_le_one _ = _ := by norm_num _ = _ := by rw [← Finset.sum_mul, ← Finset.mul_sum, ← mul_assoc, ← mul_assoc, mul_assoc _ (_ ^ _) (_ ^ _), ← pow_add, show 4 * a + (7 * a + 2) = 11 * a + 2 by lia] congr; rw [← lintegral_biUnion_finset _ fun _ _ ↦ coeGrid_measurable]; swap · rw [coe_toFinset]; exact pairwiseDisjoint_𝓙₆ simp_rw [mem_toFinset, union_𝓙₆ hu₁, ← lintegral_indicator coeGrid_measurable, eLpNorm_eq_lintegral_rpow_enorm_toReal two_ne_zero ENNReal.ofNat_ne_top, ENNReal.toReal_ofNat, one_div, show (2 : ℝ) = (2 : ℕ) by rfl, ENNReal.rpow_natCast, enorm_eq_self] congr! with x simp_rw [sq, ← inter_indicator_mul, inter_self]