Plain-language statement
For two distinct forest tops with nested spatial cubes, and for a comparison cube , the cutoff function is Lipschitz on at the scale of :
Exact Lean statement
lemma dist_χ_le (hu₁ : u₁ ∈ t) (hu₂ : u₂ ∈ t) (hu : u₁ ≠ u₂)
(h2u : 𝓘 u₁ ≤ 𝓘 u₂) (hJ : J ∈ 𝓙₅ t u₁ u₂) (mx : x ∈ 𝓘 u₁) (mx' : x' ∈ 𝓘 u₁) :
dist (χ t u₁ u₂ J x) (χ t u₁ u₂ J x') ≤ C7_5_2 a * dist x x' / D ^ s JFormal artifact
Lean source
lemma dist_χ_le (hu₁ : u₁ ∈ t) (hu₂ : u₂ ∈ t) (hu : u₁ ≠ u₂) (h2u : 𝓘 u₁ ≤ 𝓘 u₂) (hJ : J ∈ 𝓙₅ t u₁ u₂) (mx : x ∈ 𝓘 u₁) (mx' : x' ∈ 𝓘 u₁) : dist (χ t u₁ u₂ J x) (χ t u₁ u₂ J x') ≤ C7_5_2 a * dist x x' / D ^ s J := by classical by_cases hxx : x ∉ ball (c J) (8 * D ^ s J) ∧ x' ∉ ball (c J) (8 * D ^ s J) · have n₁ := χ_le_indicator hJ (x := x) rw [indicator_of_notMem hxx.1, nonpos_iff_eq_zero] at n₁ have n₂ := χ_le_indicator hJ (x := x') rw [indicator_of_notMem hxx.2, nonpos_iff_eq_zero] at n₂ rw [n₁, n₂, dist_self]; positivity rw [not_and_or, not_notMem, not_notMem] at hxx wlog hx : x ∈ ball (c J) (8 * D ^ s J) generalizing x x' · rw [or_comm] at hxx; specialize this mx' mx hxx (hxx.resolve_right hx) rwa [dist_comm, dist_comm x' x] at this clear hxx let ctx := χtilde J u₁ x let ctx' := χtilde J u₁ x' let ax := ∑ J' ∈ 𝓙₅ t u₁ u₂, χtilde J' u₁ x let ax' := ∑ J' ∈ 𝓙₅ t u₁ u₂, χtilde J' u₁ x' have ax4 : 4 ≤ ax := (four_lt_sum_χtilde hu₁ hu₂ hu h2u mx).le have ax'4 : 4 ≤ ax' := (four_lt_sum_χtilde hu₁ hu₂ hu h2u mx').le have nax : (ax : ℝ) ≠ 0 := by exact_mod_cast (zero_lt_four.trans_le ax4).ne' have nax' : (ax' : ℝ) ≠ 0 := by exact_mod_cast (zero_lt_four.trans_le ax'4).ne' have part1 : dist (χ t u₁ u₂ J x) (χ t u₁ u₂ J x') ≤ (dist x x' / D ^ s J + ctx' * dist ax ax') / 4 := by calc _ = ‖(ctx - ctx' : ℝ) / ax - ctx' * (ax - ax') / (ax * ax')‖ := by rw [mul_sub, sub_div, sub_div, mul_div_mul_right _ _ nax', mul_comm (ax : ℝ) ax', mul_div_mul_right _ _ nax, ← sub_add, sub_right_comm, sub_add_cancel]; rfl _ ≤ dist ctx ctx' / ax + ctx' * dist ax ax' / (ax * ax') := by change _ ≤ ‖(ctx - ctx' : ℝ)‖ / ax + ctx' * ‖(ax - ax' : ℝ)‖ / (ax * ax') conv_rhs => enter [1]; rw [← NNReal.norm_eq ax] conv_rhs => enter [2]; rw [← NNReal.norm_eq ctx'] enter [2]; rw [← NNReal.norm_eq ax, ← NNReal.norm_eq ax'] rw [← norm_mul, ← norm_mul, ← norm_div, ← norm_div] exact nnnorm_sub_le .. _ ≤ dist ctx ctx' / 4 + ctx' * dist ax ax' / (1 * 4) := by gcongr <;> norm_cast; exact le_trans (by norm_num) ax4 _ ≤ _ := by rw [one_mul, ← add_div]; gcongr; exact dist_χtilde_le mx mx' apply part1.trans by_cases hx' : x' ∉ ball (c J) (8 * D ^ s J) · have : ctx' = 0 := by simp_rw [ctx', ← not_ne_iff, ← zero_lt_iff, χtilde_pos_iff]; tauto rw [this, NNReal.coe_zero, zero_mul, add_zero, div_eq_inv_mul _ 4, mul_div_assoc]; gcongr rw [show 4⁻¹ = (2 : ℝ) ^ (-2 : ℝ) by norm_num, C7_5_2, NNReal.coe_pow, NNReal.coe_ofNat, ← Real.rpow_natCast, Real.rpow_le_rpow_left_iff one_lt_two] apply le_trans ?_ (by positivity) linarith rw [not_notMem] at hx' calc _ ≤ (dist x x' / D ^ s J + 8 * ∑ J' ∈ 𝓙₅ t u₁ u₂, dist (χtilde J' u₁ x) (χtilde J' u₁ x')) / 4 := by gcongr · exact χtilde_le_eight · have := dist_sum_sum_le (𝓙₅ t u₁ u₂).toFinset (fun J' ↦ (χtilde J' u₁ x : ℝ)) (fun J' ↦ χtilde J' u₁ x') exact_mod_cast this _ = (dist x x' / D ^ s J + 8 * ∑ J' ∈ 𝓙₅ t u₁ u₂ with ¬Disjoint (ball (c J) (8 * D ^ s J)) (ball (c J') (8 * D ^ s J')), dist (χtilde J' u₁ x) (χtilde J' u₁ x')) / 4 := by congr 3; refine (Finset.sum_filter_of_ne fun J' mJ' hd ↦ ?_).symm; contrapose! hd have h₁ : χtilde J' u₁ x = 0 := by have := disjoint_left.mp hd hx rw [← not_ne_iff, ← zero_lt_iff, χtilde_pos_iff]; tauto have h₂ : χtilde J' u₁ x' = 0 := by have := disjoint_left.mp hd hx' rw [← not_ne_iff, ← zero_lt_iff, χtilde_pos_iff]; tauto rw [h₁, h₂, dist_self] _ ≤ (dist x x' / D ^ s J + 8 * ∑ J' ∈ 𝓙₅ t u₁ u₂ with ¬Disjoint (ball (c J) (8 * D ^ s J)) (ball (c J') (8 * D ^ s J')), dist x x' * D ^ (1 - s J)) / 4 := by gcongr with J' mJ'; trans dist x x' / D ^ (s J') · exact dist_χtilde_le mx mx' · rw [div_eq_mul_inv, ← zpow_neg]; gcongr · exact one_le_realD a · rw [Finset.mem_filter, mem_toFinset] at mJ' rw [neg_le, neg_sub]; exact moderate_scale_change hJ mJ'.1 mJ'.2 _ = dist x x' / (D : ℝ) ^ s J * (1 / 4 + 2 * D * {J' ∈ (𝓙₅ t u₁ u₂).toFinset | ¬Disjoint (ball (c J) (8 * D ^ s J)) (ball (c J') (8 * D ^ s J'))}.card) := by rw [Finset.sum_const, nsmul_eq_mul, zpow_sub₀ (by simp), zpow_one, show 8 * (_ * (dist x x' * (D / D ^ s J))) = dist x x' / D ^ s J * (2 * D * _) * 4 by ring, add_div, mul_div_cancel_right₀ _ four_ne_zero, div_eq_mul_one_div, ← mul_add] _ ≤ _ := by rw [mul_comm, mul_div_assoc]; gcongr; exact quarter_add_two_mul_D_mul_card_le hJ- Project
- Carleson formalization
- License
- Apache-2.0
- Commit
- 74ef907d6bdb
- Source
- Carleson/ForestOperator/LargeSeparation.lean:384-471
Reuse this declaration
Bring the exact result into your workflow
The import identifies the source module. Your project still needs the pinned package dependency shown on this page.
What this badge means
This completion status comes from the project or community source. It has not yet been represented here as an independent rebuild and axiom audit.
Continue in this project
Related declarations
Adjoint Carleson adjoint
adjointCarleson_adjoint
Plain-language statement
adjointCarleson is the adjoint of carlesonOn.
Source project: Carleson formalization
Person-level attribution pending.
Ae tendsto zero of distribution le
ae_tendsto_zero_of_distribution_le
Plain-language statement
Suppose that, for every error threshold and every measure tolerance , one can choose so that the set where exceeds has measure at most . Then converges to for almost every .
Source project: Carleson formalization
Person-level attribution pending.
Antichain operator
antichain_operator
Plain-language statement
For an antichain of pairwise incomparable tiles, and measurable functions and bounded by the indicators of and , the pairing of with the Carleson sum over is controlled by the norms of and and by positive powers of the two tile-density parameters. Concretely, the bound is
Source project: Carleson formalization
Person-level attribution pending.