Two harmonic losses in the omitted-tilt majorant

Proved

Erdos390.Full.OmittedTiltHarmonicRate.tendsto_omittedTiltMajorant_mul_logL_sq_zero

by doctosil · Sep 17, 2026 · Mathlib c5ea003 (Lean v4.30.0)

analytic-number-theoryerdos-390erdos390-source-construction

Write $L=\log n$ , $y=n^{2/9}$ , $Y=\lfloor y\rfloor$ , $U=9/2$ , and $v_d=U-\log d/\log y$ . Let $\rho$ denote the Dickman function. Fix $B\ge0$ , $C,c>0$ and an integer $W>1$ . Define the omitted-tilt normalizer majorant by

M_n=\exp\!\left(\frac B{L}\frac{\log\lfloor Cn\rfloor}{\log W}\right)\frac B{L}\frac2c\sum_{W<p\le Y\atop p\ {\rm prime}}\frac1p.

M_n(\log L)^2\longrightarrow0.

The second harmonic margin is needed when an aggregated row is divided by a scale comparable to the reciprocal of $\log L$ .

Preamble

import Mathlib
import Definitions.Def_erdos390_source_bank_interface
import Definitions.Def_erdos390_dickman_step_function
import Theorems.Thm_MediumPNT
import Definitions.Def_erdos390_analytic_foundations_001
import Theorems.Thm_Erdos390_Full_DickmanBasic_kernel_product_bound
import Theorems.Thm_Erdos390_Full_DickmanBasic_rho_pos_on_zero_five
import Theorems.Thm_Erdos390_Full_DickmanBasic_kernel_secondDerivative_first_bound

Formal statement

theorem Erdos390.Full.OmittedTiltHarmonicRate.tendsto_omittedTiltMajorant_mul_logL_sq_zero :
∀ (B C c : Real) (W : Nat),
  LE.le.{0} (0 : Real) B →
    LT.lt.{0} (0 : Real) C →
      LT.lt.{0} (0 : Real) c →
        LT.lt.{0} (1 : Nat) W →
          Filter.Tendsto.{0, 0}
            (fun n =>
              HMul.hMul.{0, 0, 0} (Erdos390.Full.OmittedTiltMarkedProbability.omittedTiltMajorant B C c W n)
                (HPow.hPow.{0, 0, 0} (Real.log (Erdos390.Full.Scale.L n)) (2 : Nat)))
            Filter.atTop.{0} (nhds.{0} (0 : Real)) := by sorry

Source

https://github.com/ShouqiaoW/erdos/blob/61325b10bbdc29f4fb5e0618b414b9f2189333ad/390/lean/Erdos390/Full/OmittedTiltHarmonicRate.lean#L160-L282

View graph