Finite-shell weighted integral bound from cumulative energy
ProvedHelfgott.finite_shell_weighted_integral_boundgoldbachmeasure-theoryminor-arcssummation-by-parts
Let be a measure space and let be pairwise disjoint measurable sets. Let , assume that and are integrable, and that . Let the real weights satisfy
For real numbers , assume the cumulative energy bounds
Then
This is a finite partition version of the summation-by-parts step in Helfgott's minor-arc argument. Taking the sets to be successive arc shells, to be squared exponential-sum magnitude, and to be the magnitude of the other exponential sum allows cumulative large-sieve bounds and decreasing pointwise bounds to be combined. It does not assert any numerical energy bound or complete the Goldbach minor-arc estimate. The case is included; neither the measure space nor the total measure needs to be finite.
Preamble
import Mathlib.Algebra.BigOperators.Module import Mathlib.MeasureTheory.Integral.Bochner.Set import Mathlib.Tactic.Linarith import Mathlib.Tactic.Ring open MeasureTheory Finset open scoped BigOperators
Formal statement
theorem Helfgott.finite_shell_weighted_integral_bound {α : Type*} [MeasurableSpace α] (μ : Measure α)
(N : ℕ) (s : ℕ → Set α) (f g : α → ℝ) (w F : ℕ → ℝ)
(hs : ∀ i ≤ N, MeasurableSet (s i))
(hdisj : Set.PairwiseDisjoint (↑(range (N + 1)) : Set ℕ) s)
(hf : Integrable f μ) (hfg : Integrable (fun x => f x * g x) μ)
(hf0 : ∀ x, 0 ≤ f x)
(hg : ∀ i ≤ N, ∀ x ∈ s i, g x ≤ w i)
(hw : ∀ i < N, w (i + 1) ≤ w i) (hwN : 0 ≤ w N)
(hF : ∀ k ≤ N,
(∫ x in ⋃ i ∈ range (k + 1), s i, f x ∂μ) ≤ F k) :
(∫ x in ⋃ i ∈ range (N + 1), s i, f x * g x ∂μ) ≤
w N * F N + ∑ i ∈ range N, (w i - w (i + 1)) * F i := by sorrySource
H. A. Helfgott, The ternary Goldbach conjecture is true, arXiv:1312.7748v2, Section 6.1, Lemma 6.1 (the finite inequality immediately following equation (6.2)) and the shell decomposition in the proof of Proposition 6.2, equations (6.4)-(6.8). https://arxiv.org/html/1312.7748v2 . This formulation retains the finite Abel sum rather than replacing it with a derivative integral, and states the measure-space generality used by that step.