EMZeta certificate definitions
DefinitionArdaRH65K_Turing121_EMZetaarda-rh65k-transfernumber-theoryriemann-zeta
This bundle defines the periodic Bernoulli sawtooth function used in Euler–Maclaurin summation. Let be the set of declarations it provides. Among its named definitions are
The accompanying theorems establish the mathematical properties of these definitions.
Definition code
import Mathlib.Analysis.SpecialFunctions.ImproperIntegrals
import Mathlib.Analysis.SpecialFunctions.Pow.Deriv
import Mathlib.MeasureTheory.Function.Floor
import Mathlib.MeasureTheory.Integral.IntegralEqImproper
import Mathlib.MeasureTheory.Integral.IntervalIntegral.IntegrationByParts
import Mathlib.NumberTheory.ZetaValues
/- EMZeta.lean -- A2 Theorem 1: the Euler-Maclaurin representation of ζ with explicit
remainder.
Euler-Maclaurin summation is ABSENT from Mathlib (confirmed: `find Mathlib -iname
'*maclaurin*'` is empty; `Mathlib/Analysis/SumIntegralComparisons.lean` has only crude
monotone sum/integral bounds). This file builds it bottom-up and is upstreamable.
Architecture (proven bottom-up; only fully-proven lemmas live here, WIP in EMZetaWip.lean):
A. Periodized ("saw") Bernoulli `sawBernoulli k x := bernoulliFun k (Int.fract x)`.
Reuses `Mathlib.NumberTheory.ZetaValues.bernoulliFun` and its derivative/integral
API (`hasDerivAt_bernoulliFun`, `integral_bernoulliFun`, `contDiff_bernoulliFun`).
B. One-step Euler-Maclaurin over `[m, m+1]` for a C^1 function f:
∫ f = (f m + f (m+1))/2 - ∫ B̃₁(x) f'(x) dx
via integration by parts (`intervalIntegral.integral_mul_deriv_eq_deriv_mul`)
with the saw-1 antiderivative. This is `em_unit_step`.
C. Summed over `m = 0 .. N-1`: the classical first-order Euler-Maclaurin formula
∑_{n=1}^{N} f n = ∫_0^N f + (f N - f 0)/2 + ∫_0^N B̃₁(x) f'(x) dx
(`euler_maclaurin_one`). Telescoping of the integral (`intervalIntegral.sum_integral_
adjacent_intervals`) + telescoping of the trapezoid endpoints.
D. Specialisation to `f x = x^{-s}` on `[1, N]` (real `s`) and the tail limit `N → ∞`
giving the K=1 Euler-Maclaurin representation of `ζ` for real `s > 1`
(`em_zeta_real_K1`), matching the Mathlib fractional-part representation route
used in `Mathlib.NumberTheory.Harmonic.ZetaAsymp`.
NOT YET in this file (see EMZetaWip.lean, never imported by a guard, never `sorry` here):
- the general order-K induction (periodized `bernoulliFun k` for k ≥ 2);
- the complex interval-σ form valid `Re s > 1 - 2K` by analytic continuation
(`AnalyticOnNhd.eqOn`);
- the explicit remainder BOUND `|R_K| ≤ (|B_{2K}|/(2K)!) ∫ |f^{(2K)}|`.
conjecture1_proved = False. This is NOT a proof of RH; it is a classical analysis lemma.
-/
open MeasureTheory intervalIntegral Set Filter Topology
open scoped Real
namespace ZetaReflection
/-! ## A. The periodized ("saw") Bernoulli functions -/
/-- The periodized Bernoulli function `B̃ₖ(x) = Bₖ({x})` on `ℝ`, where `{x} = Int.fract x`.
For `k = 1` this is the classical sawtooth `{x} - 1/2` (away from integers). We build EM
on this `ℝ → ℝ` form rather than Mathlib's `AddCircle`-valued `periodizedBernoulli`, since
the finite-sum induction wants a plain real function with `bernoulliFun`'s derivative and
integral API directly available. -/
noncomputable def sawBernoulli (k : ℕ) (x : ℝ) : ℝ := bernoulliFun k (Int.fract x)
/-! ## B. One-step Euler-Maclaurin over a unit cell
For a `C¹` function `f` on `[m, m+1]` (real, integer `m`), integration by parts with the
order-1 antiderivative `w x = x - m - 1/2` of the constant `1` (equal to `sawBernoulli 1`
on the cell) yields the trapezoidal Euler-Maclaurin step
∫_m^{m+1} f = (f m + f(m+1))/2 − ∫_m^{m+1} (sawBernoulli 1 x) · f'(x) dx.
-/
/-! ## C. Summed first-order Euler-Maclaurin over `[0, N]`
Summing `em_unit_step` over the cells `[0,1], [1,2], …, [N-1,N]` and telescoping the two
integrals with `intervalIntegral.sum_integral_adjacent_intervals` gives the classical
first-order Euler-Maclaurin formula (sum indexed `1..N`, integral over `[0,N]`)
∑_{n=1}^{N} f n = ∫_0^N f + (f N − f 0)/2 + ∫_0^N (sawBernoulli 1 x)·f'(x) dx.
(Equivalently `∑_{n=0}^{N} f n = ∫_0^N f + (f 0 + f N)/2 + remainder`, the textbook form.)
-/
/-! ## D. Zeta base case: first-order Euler-Maclaurin for `f x = x^{-s}` on `[1, N]`
The reusable finite-`N` statement. Specialising the general `[M, N]` machinery below to
`M = 1`, `f x = x^{-s}` with derivative `f' x = -s · x^{-s-1}` (valid for `x ≠ 0`, hence on
all of `[1, N]`), we get the classical finite Euler-Maclaurin representation of the partial
zeta sum. Taking `N → ∞` (real `s > 1`) matches the fractional-part representation in
`Mathlib.NumberTheory.Harmonic.ZetaAsymp` (`zeta_limit_aux1`, `termTSum`).
-/
end ZetaReflection
def ArdaRH65K_Turing121_EMZeta : Nat := 0
Source