The Vitali convergence theorem
ProvedFamousTheorems.vitali_convergenceThe Vitali convergence theorem. Let and let . Then in if and only if in measure and the family is uniformly integrable and uniformly tight in .
It characterises convergence exactly, and it strengthens the dominated convergence theorem, whose hypothesis is only sufficient. In probability it gives the standard criterion for convergence of moments: convergence in probability plus uniform integrability.
Formalization note. Mathlib's MeasureTheory.tendstoInMeasure_iff_tendsto_Lp, for an arbitrary (not necessarily finite) measure. This is why uniform tightness (UnifTight) appears alongside uniform integrability (UnifIntegrable). convergence is expressed by eLpNorm (f n - g) p μ → 0.
Preamble
import Mathlib
Formal statement
namespace FamousTheorems
theorem vitali_convergence {α β : Type*} {m : MeasurableSpace α} [NormedAddCommGroup β] {μ : MeasureTheory.Measure α} {p : ENNReal}
{f : ℕ → α → β} {g : α → β} (hp : 1 ≤ p) (hp' : p ≠ ⊤) (hf : ∀ n, MeasureTheory.MemLp (f n) p μ)
(hg : MeasureTheory.MemLp g p μ) :
MeasureTheory.TendstoInMeasure μ f Filter.atTop g ∧ MeasureTheory.UnifIntegrable f p μ ∧
MeasureTheory.UnifTight f p μ ↔
Filter.Tendsto (fun n => MeasureTheory.eLpNorm (f n - g) p μ) Filter.atTop (nhds 0) := by sorry
end FamousTheoremsSource
Listed in Mathlib's curated theorem manifests (docs/1000.yaml); formalized in Mathlib as `MeasureTheory.tendstoInMeasure_iff_tendsto_Lp`. Proof here reduces to that Mathlib result.