Jensen's inequality (integral form)
ProvedFamousTheorems.jensen_inequality_integralJensen's inequality (integral form). Let be a probability measure, a closed convex subset of a real Banach space , and convex and continuous on . If takes values in almost everywhere and both and are integrable, then
It is one of the most used inequalities in analysis and probability. Special cases include the AM–GM inequality, , and the nonnegativity of relative entropy, and it underlies the monotonicity of norms on probability spaces.
Formalization note. Mathlib's ConvexOn.map_integral_le. Integrals are Bochner integrals, and ∀ᵐ x ∂μ, f x ∈ s means for -almost every .
Preamble
import Mathlib
Formal statement
namespace FamousTheorems
theorem jensen_inequality_integral {α E : Type*} {m0 : MeasurableSpace α} [NormedAddCommGroup E] [NormedSpace ℝ E] [CompleteSpace E]
{μ : MeasureTheory.Measure α} [MeasureTheory.IsProbabilityMeasure μ] {s : Set E} {f : α → E} {g : E → ℝ}
(hg : ConvexOn ℝ s g) (hgc : ContinuousOn g s) (hsc : IsClosed s) (hfs : ∀ᵐ x ∂μ, f x ∈ s)
(hfi : MeasureTheory.Integrable f μ) (hgi : MeasureTheory.Integrable (g ∘ f) μ) :
g (∫ x, f x ∂μ) ≤ ∫ x, g (f x) ∂μ := by sorry
end FamousTheoremsSource
Listed in Mathlib's curated theorem manifests (docs/1000.yaml); formalized in Mathlib as `ConvexOn.map_integral_le`. Proof here reduces to that Mathlib result.