Ported for Lean Pool: imports relocated and code adapted to Lean/Mathlib v4.34.0.
Statement of the convex Aleksandrov theorem #
This file freezes the target theorem surface without asserting the theorem as an axiom.
The target Aleksandrov second order differentiability statement for convex functions.
The statement is intentionally formulated over a finite-dimensional real inner product space, rather than a matrix space. The almost-everywhere assertion is taken with respect to Lebesgue measure restricted to the open convex domain.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equivalent null-exceptional-set formulation of ConvexAleksandrovAEStatement.
Equations
Instances For
Null-exceptional-set formulation using the exceptional set already intersected with the domain.
Equations
Instances For
Fubini/slicing target for the finite-dimensional proof: for every fixed direction, the set of base points whose affine-line restriction has scalar quadratic estimate data at the base parameter has full measure in the domain.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
DirectionalLineScalarEstimateFubiniStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Unit-direction version of DirectionalLineScalarEstimateFubiniStatement. Scalar-rescaling
invariance and the zero-direction lemma turn this into the all-directions statement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
DirectionalLineScalarUnitFubiniStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
The stronger local scalar directional-line Fubini target implies its punctured normalized quotient version.
Unit-direction scalar Fubini implies the all-directions scalar Fubini target.
Unit-direction quotient Fubini implies the all-directions quotient Fubini target.
The stronger local scalar unit-direction Fubini target implies its punctured normalized quotient version.
Slicewise version of DirectionalLineScalarEstimateFubiniStatement: for every affine line
parallel to a fixed direction, almost every line parameter gives an ambient base point carrying
scalar estimate data in that same direction. This is the one-dimensional input that a future
finite-dimensional Fubini/disintegration argument should integrate over transverse parameters.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
DirectionalLineScalarEstimateSlicewiseStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
The stronger local scalar slicewise target implies its punctured normalized quotient version.
The remaining finite-dimensional measure-theoretic transfer for the directional-line scalar
estimate target: slicewise almost-everywhere goodness along every affine line parallel to ξ
implies ambient full measure of the fixed-direction good set.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
DirectionalLineScalarSlicewiseToFubiniStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Unit-direction version of DirectionalLineScalarSlicewiseToFubiniStatement. Since scalar
directional-line good sets are invariant under nonzero rescaling and the zero direction is
automatic, this is enough for the all-directions transfer.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Unit-direction quotient version of DirectionalLineScalarSlicewiseToFubiniStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A unit-direction slicewise-to-Fubini transfer gives the all-directions transfer.
Quotient version of
DirectionalLineScalarUnitSlicewiseToFubiniStatement.fubiniStatement.
Finite-family version of the slicewise-to-Fubini transfer tailored to the polarization
argument. It asks only for ambient full measure in the selected pure directions v i and in the
off-diagonal pairwise-sum directions v i + v j.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
FiniteDirectionalLineScalarSlicewiseToFullMeasureStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Compatible finite-slice full-measure target: for a chosen finite direction family, the set of base points carrying scalar line estimates with one shared ambient first-order slope has full measure in the domain. This is stronger than the raw linewise Fubini target and is the natural input for compatible finite-dimensional reconstruction.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
CompatibleDirectionalSliceFullMeasureStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Symmetric-compatible finite-slice full-measure target. This is the same compatible first-order finite-slice data, with pairwise-sum quadratic coefficients already symmetrized on the finite frame.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Punctured normalized quotient version of
SymmetricCompatibleDirectionalSliceFullMeasureStatement.
Equations
- One or more equations did not get rendered due to their size.
Instances For
The non-quotient compatible finite-slice full-measure target implies its quotient version.
Compatible full-measure slice data implies full measure of the symmetrized compatible finite-slice locus.
Compatible quotient full-measure slice data implies full measure of the symmetrized compatible quotient finite-slice locus.
Non-quotient compatible full-measure slice data implies full measure of the symmetrized compatible quotient finite-slice locus.
Symmetric-compatible full-measure slice data implies its punctured normalized quotient version with the same symmetric finite-slice data.