Lebesgue points along concentric closed balls #
Mathlib's IsUnifLocDoublingMeasure.ae_tendsto_average_norm_sub states the Lebesgue
differentiation theorem for a uniformly locally doubling measure along arbitrary families of
closed balls shrinking to a point. This file records its most common special case: almost every
point x is a Lebesgue point of a locally integrable function f in the sense that the mean
oscillation ⨍_{closedBall x r} ‖f - f x‖ tends to zero as r → 0⁺, and the analogous statement
for the averages of an integrable ℝ≥0∞-valued function. For an additive Haar measure on a
finite-dimensional real normed space, spheres are null, so the same averages over open balls
converge as well.
Lebesgue differentiation theorem along concentric closed balls. For a uniformly locally
doubling measure, almost every point x is a Lebesgue point of a locally integrable function f:
the mean oscillation ⨍_{closedBall x r} ‖f - f x‖ tends to zero as r → 0⁺.
Lebesgue differentiation theorem along concentric closed balls, for ℝ≥0∞-valued
functions. For a uniformly locally doubling measure and an a.e. measurable f with finite
integral, the averages ⨍_{closedBall x r} f tend to f x as r → 0⁺ for almost every x.
Lebesgue differentiation theorem along concentric open balls, for an additive Haar
measure. For an a.e. measurable f : E → ℝ≥0∞ with finite integral, the averages
⨍_{ball x r} f tend to f x as r → 0⁺ for almost every x. Open and closed balls of positive
radius differ by a null set, so this is IsUnifLocDoublingMeasure.ae_tendsto_setLAverage_closedBall
for open balls.