LeanAide GPT-4 documentation

Module: Mathlib.MeasureTheory.Function.ConditionalExpectation.Unique



MeasureTheory.lintegral_nnnorm_le_of_forall_fin_meas_integral_eq

theorem MeasureTheory.lintegral_nnnorm_le_of_forall_fin_meas_integral_eq : ∀ {α : Type u_1} {m m0 : MeasurableSpace α} {μ : MeasureTheory.Measure α} {s : Set α}, m ≤ m0 → ∀ {f g : α → ℝ}, MeasureTheory.StronglyMeasurable f → MeasureTheory.IntegrableOn f s μ → MeasureTheory.StronglyMeasurable g → MeasureTheory.IntegrableOn g s μ → (∀ (t : Set α), MeasurableSet t → ↑↑μ t < ⊤ → ∫ (x : α) in t, g x ∂μ = ∫ (x : α) in t, f x ∂μ) → MeasurableSet s → ↑↑μ s ≠ ⊤ → ∫⁻ (x : α) in s, ↑‖g x‖₊ ∂μ ≤ ∫⁻ (x : α) in s, ↑‖f x‖₊ ∂μ

The theorem `MeasureTheory.lintegral_nnnorm_le_of_forall_fin_meas_integral_eq` states that for a given sub-σ-algebra `m` of `m0`, and two functions `f` and `g` which are measurable in `m0` and `m` respectively, if their integrals over any `m`-measurable set with finite measure are the same, then the integral over an `m`-measurable set `s` with finite measure of the nonnegative norm of `g` is less than or equal to that of `f`. This condition holds if `f` and `g` are both strongly measurable, and are integrable over the set `s` with respect to measure `μ`.

More concisely: If two measurable functions `f` and `g` have equal integrals over every `m`-measurable set of finite measure in `μ`, and `f` and `g` are strongly measurable and integrable over an `m`-measurable set `s` with finite measure, then the integral of the nonnegative norm of `g` over `s` is less than or equal to that of `f`.

MeasureTheory.integral_norm_le_of_forall_fin_meas_integral_eq

theorem MeasureTheory.integral_norm_le_of_forall_fin_meas_integral_eq : ∀ {α : Type u_1} {m m0 : MeasurableSpace α} {μ : MeasureTheory.Measure α} {s : Set α}, m ≤ m0 → ∀ {f g : α → ℝ}, MeasureTheory.StronglyMeasurable f → MeasureTheory.IntegrableOn f s μ → MeasureTheory.StronglyMeasurable g → MeasureTheory.IntegrableOn g s μ → (∀ (t : Set α), MeasurableSet t → ↑↑μ t < ⊤ → ∫ (x : α) in t, g x ∂μ = ∫ (x : α) in t, f x ∂μ) → MeasurableSet s → ↑↑μ s ≠ ⊤ → ∫ (x : α) in s, ‖g x‖ ∂μ ≤ ∫ (x : α) in s, ‖f x‖ ∂μ

The theorem states that given a sub-measure space `m` which is a subset of `m0`, a function `f` that is measurable with respect to `m0`, and a function `g` that is measurable with respect to `m`, if the integrals of these two functions coincide on every `m`-measurable set with finite measure, then the integral over any `m`-measurable set `s` with finite measure of the norm of `g` is less than or equal to the integral over the same set of the norm of `f`. This is under the condition that both `f` and `g` are strongly measurable and integrable on `s`.

More concisely: Given measurable and integrable functions `f` and `g` on a sub-measure space `m` of `m0`, if the integrals of their norms coincide on every `m`-measurable set with finite measure, then the integral of `g`'s norm is less than or equal to the integral of `f`'s norm on any `m`-measurable set with finite measure.

MeasureTheory.Lp.ae_eq_of_forall_set_integral_eq'

theorem MeasureTheory.Lp.ae_eq_of_forall_set_integral_eq' : ∀ {α : Type u_1} {E' : Type u_2} (𝕜 : Type u_4) {p : ENNReal} {m m0 : MeasurableSpace α} {μ : MeasureTheory.Measure α} [inst : RCLike 𝕜] [inst_1 : NormedAddCommGroup E'] [inst : InnerProductSpace 𝕜 E'] [inst : CompleteSpace E'] [inst : NormedSpace ℝ E'], m ≤ m0 → ∀ (f g : ↥(MeasureTheory.Lp E' p μ)), p ≠ 0 → p ≠ ⊤ → (∀ (s : Set α), MeasurableSet s → ↑↑μ s < ⊤ → MeasureTheory.IntegrableOn (↑↑f) s μ) → (∀ (s : Set α), MeasurableSet s → ↑↑μ s < ⊤ → MeasureTheory.IntegrableOn (↑↑g) s μ) → (∀ (s : Set α), MeasurableSet s → ↑↑μ s < ⊤ → ∫ (x : α) in s, ↑↑f x ∂μ = ∫ (x : α) in s, ↑↑g x ∂μ) → MeasureTheory.AEStronglyMeasurable' m (↑↑f) μ → MeasureTheory.AEStronglyMeasurable' m (↑↑g) μ → μ.ae.EventuallyEq ↑↑f ↑↑g

This theorem, referred to as the "Uniqueness of the conditional expectation", states that: Given - two types `α` and `E'`, - a type `𝕜`, - an extended nonnegative real number `p`, - two measurable spaces `m` and `m0` on `α`, - a measure `μ` on `α`, - and assuming `𝕜` is a real-like type, `E'` is a normed additive commutative group, is an inner product space over `𝕜`, is a complete space, and is a normed space over the real numbers, - and also assuming that the measurable space `m` is less than or equal to `m0`, for any two functions `f` and `g` from the Lp space on `E'` with parameter `p` and measure `μ`, if - `p` is non-zero and finite, - both `f` and `g` are integrable over any measurable set `s` with finite measure `μ`, - the integral of `f` and `g` over any such set `s` are equal, - both `f` and `g` are almost everywhere strongly measurable with respect to the measure `μ` (in the sense that they are almost everywhere equal to a strongly measurable function), then `f` and `g` are equal almost everywhere with respect to the measure `μ`. This theorem thus provides a uniqueness property for functions satisfying these conditions.

More concisely: If two functions from the Lp space over a measurable space with finite measure, are integrable, equal in integral over every measurable set with finite measure, and almost everywhere strongly measurable, then they are equal almost everywhere.