OAI/NumberTheory/EgyptianFractions/ThreePrimeSingularTail.lean
11 declarations, 0 contain sorry, 0 declared as axioms
- lemma
OAI.Problem337.shifted_reciprocal_square_tailline 11 - theorem
OAI.Problem337.threePrimeOddFactors_tail_abs_sub_oneline 31Source documentation
Every finite tail of the canonical ternary singular series has relative error at most
2 / (N + 1), uniformly in the integer being represented.theorem threePrimeOddFactors_tail_abs_sub_one (u N : ℕ) (P : Finset ℕ) (hP : ∀ n ∈ P, N ≤ n) : |(∏ n ∈ P, (1 + threePrimeOddPerturbation u n)) - 1| ≤ 2 / ((N : ℝ) + 1) := by …Mentions: OAI.Problem337.threePrimeOddPerturbation
- theorem
OAI.Problem337.threePrimeOddFactors_partial_abs_leline 53Source documentation
Uniform boundedness of the finite odd-prime factors.
theorem threePrimeOddFactors_partial_abs_le (u N : ℕ) : |∏ n ∈ Finset.range N, (1 + threePrimeOddPerturbation u n)| ≤ 3 := by …Mentions: OAI.Problem337.threePrimeOddPerturbation
- theorem
OAI.Problem337.threePrimeOddFactors_tprod_sub_partialline 66Source documentation
The canonical infinite odd-prime product is uniformly approximated by its first
Nfactors at the explicit rate6 / (N + 1).theorem threePrimeOddFactors_tprod_sub_partial (u N : ℕ) : |(∏' n : ℕ, (1 + threePrimeOddPerturbation u n)) - ∏ n ∈ Finset.range N, (1 + threePrimeOddPerturbation u n)| ≤ 6 / ((N : ℝ) + 1) := by …Mentions: OAI.Problem337.threePrimeOddPerturbation
- def
OAI.Problem337.threePrimeSingularPartialline 88Source documentation
The finite canonical singular product, including the factor at two.
def threePrimeSingularPartial (u N : ℕ) : ℝ := …
- theorem
OAI.Problem337.threePrimeSingularSeries_sub_partialline 93Source documentation
Uniform quantitative truncation of the actual defined singular series.
theorem threePrimeSingularSeries_sub_partial (u N : ℕ) : |threePrimeSingularSeries u - threePrimeSingularPartial u N| ≤ 12 / ((N : ℝ) + 1) := by …Mentions: OAI.Problem337.threePrimeSingularSeries, OAI.Problem337.threePrimeSingularPartial
- theorem
OAI.Problem337.threePrimeSingularPartial_tendstoUniformlyline 106Source documentation
Uniform convergence in the represented integer: the onset is independent of
u, including even inputs where the singular series vanishes.theorem threePrimeSingularPartial_tendstoUniformly : TendstoUniformly (fun N u => threePrimeSingularPartial u N) threePrimeSingularSeries atTop := by …Mentions: OAI.Problem337.threePrimeSingularPartial, OAI.Problem337.threePrimeSingularSeries
- theorem
OAI.Problem337.threePrimeSingularSeries_abs_le_sixline 123Source documentation
A simultaneous absolute bound for all singular series, useful when converting an additive asymptotic error into a relative error.
theorem threePrimeSingularSeries_abs_le_six (u : ℕ) : |threePrimeSingularSeries u| ≤ 6 := by …Mentions: OAI.Problem337.threePrimeSingularSeries
- theorem
OAI.Problem337.threePrimeSingularPartial_sub_tendsto_zeroline 135Source documentation
A cutoff tending to infinity may depend on the represented integer or on any other parameter; uniform truncation still makes the error tend to zero.
theorem threePrimeSingularPartial_sub_tendsto_zero {α : Type*} {l : Filter α} (u Q : α → ℕ) (hQ : Tendsto Q l atTop) : Tendsto (fun x => threePrimeSingularSeries (u x) - threePrimeSingularPartial (u x) (Q x)) l (nhds 0) := by …Mentions: OAI.Problem337.threePrimeSingularSeries, OAI.Problem337.threePrimeSingularPartial
- theorem
OAI.Problem337.threePrimeSingularPartial_weighted_isLittleOline 147Source documentation
Truncating the singular series changes any weighted main term by little-o of that same weight. No positivity or nonvanishing of the weight is needed.
theorem threePrimeSingularPartial_weighted_isLittleO {α : Type*} {l : Filter α} (u Q : α → ℕ) (hQ : Tendsto Q l atTop) (w : α → ℝ) : (fun x => (threePrimeSingularSeries (u x) - threePrimeSingularPartial (u x) (Q x)) * w x) =o[l] w := by …Mentions: OAI.Problem337.threePrimeSingularSeries, OAI.Problem337.threePrimeSingularPartial
- theorem
OAI.Problem337.threePrimeSingularPartial_asymptotic_iffline 159Source documentation
An asymptotic formula with a growing finite singular product is equivalent to the formula with the canonical infinite series. This supplies the uniform Euler-tail transfer in a major-arc computation without asserting that formula.
theorem threePrimeSingularPartial_asymptotic_iff {α : Type*} {l : Filter α} (u Q : α → ℕ) (hQ : Tendsto Q l atTop) (a w : α → ℝ) : ((fun x => a x - threePrimeSingularSeries (u x) * w x) =o[l] w) ↔ ((fun x => a x - threePrimeSingularPartial (u x) (Q x) * w x) =o[l] w) := by …Mentions: OAI.Problem337.threePrimeSingularSeries, OAI.Problem337.threePrimeSingularPartial